Differential SOD2 and GSTZ1 profiles contribute to contrasting dental pulp stem cell susceptibilities to oxidative damage and premature senescence

Background Dental pulp stem cells (DPSCs) are increasingly being advocated as viable cell sources for regenerative medicine-based therapies. However, significant heterogeneity in DPSC expansion and multi-potency capabilities are well-established, attributed to contrasting telomere profiles and susceptibilities to replicative senescence. As DPSCs possess negligible human telomerase (hTERT) expression, we examined whether intrinsic differences in the susceptibilities of DPSC sub-populations to oxidative stress-induced biomolecular damage and premature senescence further contributed to this heterogeneity, via differential enzymic antioxidant capabilities between DPSCs. Methods DPSCs were isolated from human third molars by differential fibronectin adhesion, and positive mesenchymal (CD73/CD90/CD105) and negative hematopoietic (CD45) stem cell marker expression confirmed. Isolated sub-populations were expanded in H2O2 (0–200 μM) and established as high or low proliferative DPSCs, based on population doublings (PDs) and senescence (telomere lengths, SA-β-galactosidase, p53/p16INK4a/p21waf1/hTERT) marker detection. The impact of DPSC expansion on mesenchymal, embryonic, and neural crest marker expression was assessed, as were the susceptibilities of high and low proliferative DPSCs to oxidative DNA and protein damage by immunocytochemistry. Expression profiles for superoxide dismutases (SODs), catalase, and glutathione-related antioxidants were further compared between DPSC sub-populations by qRT-PCR, Western blotting and activity assays. Results High proliferative DPSCs underwent > 80PDs in culture and resisted H2O2−induced senescence (50–76PDs). In contrast, low proliferative sub-populations exhibited accelerated senescence (4–32PDs), even in untreated controls (11-34PDs). While telomere lengths were largely unaffected, certain stem cell marker expression declined with H2O2 treatment and expansion. Elevated senescence susceptibilities in low proliferative DPSC (2–10PDs) were accompanied by increased oxidative damage, absent in high proliferative DPSCs until 45–60PDs. Increased SOD2/glutathione S-transferase ζ1 (GSTZ1) expression and SOD activities were identified in high proliferative DPSCs (10–25PDs), which declined during expansion. Low proliferative DPSCs (2–10PDs) exhibited inferior SOD, catalase and glutathione-related antioxidant expression/activities. Conclusions Significant variations exist in the susceptibilities of DPSC sub-populations to oxidative damage and premature senescence, contributed to by differential SOD2 and GSTZ1 profiles which maintain senescence-resistance/stemness properties in high proliferative DPSCs. Identification of superior antioxidant properties in high proliferative DPSCs enhances our understanding of DPSC biology and senescence, which may be exploited for selective sub-population screening/isolation from dental pulp tissues for regenerative medicine-based applications. Supplementary Information The online version contains supplementary material available at 10.1186/s13287-021-02209-9.


(Continued from previous page)
Results: High proliferative DPSCs underwent > 80PDs in culture and resisted H 2 O 2− induced senescence . In contrast, low proliferative sub-populations exhibited accelerated senescence (4-32PDs), even in untreated controls . While telomere lengths were largely unaffected, certain stem cell marker expression declined with H 2 O 2 treatment and expansion. Elevated senescence susceptibilities in low proliferative DPSC (2-10PDs) were accompanied by increased oxidative damage, absent in high proliferative DPSCs until 45-60PDs. Increased SOD2/ glutathione S-transferase ζ1 (GSTZ1) expression and SOD activities were identified in high proliferative DPSCs , which declined during expansion. Low proliferative DPSCs (2-10PDs) exhibited inferior SOD, catalase and glutathione-related antioxidant expression/activities. Conclusions: Significant variations exist in the susceptibilities of DPSC sub-populations to oxidative damage and premature senescence, contributed to by differential SOD2 and GSTZ1 profiles which maintain senescenceresistance/stemness properties in high proliferative DPSCs. Identification of superior antioxidant properties in high proliferative DPSCs enhances our understanding of DPSC biology and senescence, which may be exploited for selective sub-population screening/isolation from dental pulp tissues for regenerative medicine-based applications.

Background
Dental pulp stem cells (DPSCs) are increasingly being advocated as a viable stem cell source in the development of regenerative medicine-based therapies, based on their accessibility, self-renewal, clonogenicity, and multipotency [1][2][3]. However, a drawback associated with DPSC-based therapy development is their significant heterogeneity within dental pulp tissues, with individual clones demonstrating major differences in proliferation and differentiation capabilities [4][5][6]. Consequently, despite heterogeneous DPSC populations achieving > 120 population doublings (PDs) in vitro, only 20% of purified DPSCs are capable of proliferating > 20PDs. Such issues are confounded by DPSCs being proposed to exist within distinct niches within dental pulp tissues (subodontoblast layer, pulpal vasculature and central pulp) [7], which increases their complexity regarding the origins and regenerative characteristics of individual DPSC sub-populations. Such features have major implications for successful DPSC exploitation, as a significant limitation of mesenchymal stem cell (MSC)-based therapies is the extensive in vitro expansion necessary to produce sufficient cell numbers for clinical use. Consequently, cell expansion eventually leads to proliferative decline and replicative (telomere-dependent) senescence, characterized by progressive telomere shortening, inhibition of G 1 -S phase transition and permanent growth arrest. This is associated with the loss of telomeric TTAGGG repeats, positive senescence-associated (SA)-β-galactosidase staining, and increased tumor suppressor (p53 and retinoblastoma protein, pRb) and cyclin-dependent kinase inhibitor (p21 waf1 and p16 INK4a ) gene expression [8][9][10]. Such events are recognized to significantly alter MSC genotype and phenotype, leading to impaired regenerative properties and disrupted local tissue micro-environment signaling mechanisms, through the secretome associated with the senescence-associated secretory phenotype (SASP) [10,11]. Despite significant differences in the ex vivo expansion capabilities of individual DPSC sub-populations, only recently have studies investigated such variations in proliferative capabilities and senescence susceptibilities on the multi-potency and other properties of different DPSC sub-populations [12,13]. High proliferative DPSCs are reported to achieve > 80PDs, whereas low proliferating DPSCs only complete < 40PDs before senescence, correlating with DPSCs with high proliferative capacities possessing longer telomeres than less proliferative subpopulations. Low proliferative DPSC senescence was also associated with the loss of stem cell marker characteristics and impaired osteogenic/chondrogenic differentiation, in favor of adipogenesis. In contrast, high proliferative DPSCs retained multi-potency capabilities, only demonstrating impaired differentiation following prolonged in vitro expansion (> 60PDs). As most studies have reported no or negligible reverse transcriptase, human telomerase catalytic subunit (hTERT) expression in human DPSCs [12,[14][15][16], hTERT is unlikely to be responsible for maintaining telomere integrity and the proliferative/multi-potency capabilities of high proliferative DPSCs. Therefore, other intrinsic mechanisms may account for differences in telomere lengths, proliferation rates, and differentiation capabilities between high and low proliferative DPSC sub-populations.
In light of the evidence attributing superior endogenous enzymic antioxidant capabilities with cellular resistance to oxidative stress, we investigated whether similar differences in DPSC susceptibilities to oxidative stressinduced biomolecular damage and premature senescence existed, due to differential enzymic antioxidant capabilities between DPSC sub-populations. Consequently, this is the first study to confirm inherent differences in enzymic antioxidant expression profiles between high and low proliferative DPSC sub-populations. Such SOD2 and glutathione S-transferase ζ1 (GSTZ1) adaptations would contribute to the protection of high proliferative DPSCs from oxidative damage and senescence, thereby helping explain DPSC sub-population heterogeneity overall.

Stem cell isolation and culture under oxidative stress conditions
Human DPSCs were isolated from third molar teeth collected from patients (all female, age 18-30 years) undergoing orthodontic extractions at the School of Dentistry, Cardiff University, UK. Teeth were collected in accordance with the Declaration of Helsinki (2013), with informed patient consent and ethical approval by the South East Wales Research Ethics Committee of the National Research Ethics Service (NRES), UK.

Telomere restriction fragment (TRF) length analysis
At selected PDs throughout their proliferative lifespans, DPSCs expanded with or without H 2 O 2 (0-200 μM) were maintained in 6-well plates as above, until 80-90% confluent. Following DNA purification [12], telomere length analyses were performed using the TeloTAGGG Telomere Restriction Fragment (TRF) Length Assay Kit (Roche, Welwyn Garden City, UK), per the manufacturer's instructions. A digoxigenin (DIG)-labeled molecular weight marker (kb, in Kit) and positive DIGlabeled control DNA sample (CTRL, in Kit) were also included. Mean telomere lengths were calculated from Southern blot images via ImageJ® Software [12].

Immuno-detection of oxidative stress-induced biomarkers
DPSCs expanded with or without H 2 O 2 (0-200 μM) were assessed for the presence of oxidative DNA and protein biomarker levels, by immunocytochemistry using 8-well chamber slides (VWR International, Lutterworth, UK). Oxidative DNA damage, in the form of 8-hydroxydeoxy-guanosine (8-OHdG) levels [21], was detected using fluorometric OxyDNA Assay Kits (Merck Millipore, Watford, UK). Oxidative protein damage (in the form of protein carbonyl content [22]), was detected using fluorometric OxyICC™ Oxidized Protein Detection Kits (Merck Millipore). Control wells were included for each Kit, consisting of phosphate-buffered saline (PBS), instead of fluorescent conjugates. Chamber slides were subsequently mounted using FluorSave Reagent (Merck Millipore) and viewed using a Leica Dialux 20 Fluorescent Microscope (Leica Microsystems, Milton Keynes, UK). Images were captured using HCImage acquisition and analysis software (Hamamatsu Corporation, Sewickley, PA, USA).

Enzymic antioxidant activity analysis
DPSCs expanded with or without H 2 O 2 (0-200 μM) were harvested for assessment of total SOD, catalase, and GPX activities, per manufacturer's instructions. Total SOD activities were determined using SOD Activity Colorimetric Assay Kits (Abcam). Total catalase activities were determined using Catalase Specific Activity Assay Kits (Abcam). Total GPX activities were determined using Glutathione Peroxidase Assay Kits (Cambridge Bioscience, Cambridge, UK). Sample absorbance values were read spectrophotometrically using a using a FLUOstar® Omega Plate Reader (BMG Labtech, Aylesbury, UK) and total SOD, catalase, and GPX activities in cell extracts determined versus SOD (in Kit), catalase (from human erythrocytes, Sigma), and GPX (from human erythrocytes, Sigma) standard curves.

Statistical analysis
Each experiment was performed on n = 3 independent occasions. Statistical analyses were performed using GraphPad Prism (GraphPad Software, La Jolla, CA, USA). Data were expressed as mean ± standard error of mean (SEM) and statistically compared using analysis of variance (ANOVA), with post hoc Tukey test. Significance was considered at p < 0.05.

DPSC population doublings under oxidative stress conditions
Several DPSC sub-populations were successfully isolated and characterized from 3 individual patient donors (patients A, C, and D). As significant variations in proliferative capacity and susceptibilities to replicative senescence have previously been identified within different DPSC sub-populations [12], initial studies assessed the effects of continual sub-culture under oxidative stress (0-200 μM H 2 O 2 ) conditions on PDs throughout their proliferative lifespans to senescence for individual DPSC sub-populations. Overall, PDs and proliferative capacities showed marked variations in DPSC susceptibilities to premature senescence, with PD differences irrespective of which patient DPSCs were derived.

DPSC stem cell marker detection
As DPSC senescence is commonly associated with the loss of stem cell marker expression [12,39], RT-PCR analysis was performed to determine whether increasing H 2 O 2 treatment accelerated the loss of stem cell characteristics in DPSCs. All DPSC sub-populations showed varying positive gene expression for MSC markers, CD73, CD90, and CD105 (Fig. 4a-d). In contrast, hematopoietic stem cell marker, CD45, was undetectable in all DPSCs. CD90 expression was largely retained in all DPSC sub-populations through culture expansion, irrespective of H 2 O 2 treatment. However, CD73 and CD105 expression by high proliferative sub-population, A1, showed declined detection at 40-65PDs, dependently and independently of H 2 O 2 treatment (Fig. 4a). In contrast, reductions in CD73 and CD105 expression were less evident in low proliferative DPSCs, A2, C3, and D4 ( Fig. 4b-d, respectively).

Oxidative stress biomarker detection in DPSCs
As premature senescence is commonly associated with increased oxidative DNA and protein damage [20][21][22], we next investigated whether the high and low proliferative DPSC sub-populations also differed in their respective susceptibilities to oxidative stress-induced biomarker formation. Overall, oxidative DNA damage, in the form of 8-OHdG, showed marked variations in detection between high and low proliferative DPSC sub-populations (Fig. 6). High proliferative DPSC sub-population, A1, at 2-10PDs exhibited least positive nuclear DNA fluorescence detection for oxidative DNA damage overall, especially in untreated and 50 μM H 2 O 2 -treated cultures  Fig. 6a, i-ii). Although low-intensity cytoplasmic background staining was present, limited nuclear fluorescence was evident. However, increased nuclear DNA fluorescence staining intensities was identified for A1 at 2-10PDs, with 100-200 μM H 2 O 2 (arrowed, Fig. 6a, iiiiv). There was also strong co-localization between oxidative DNA (fluorescein isothiocyanate, FITC) and Hoechst nuclear staining (arrowed, Fig. 6a, v-viii), thereby confirming the prominent nuclear localization of the oxidative DNA damage. In contrast, low proliferative DPSC sub-populations, A2 (Fig. S3), C3 (Fig. S3), and D4 (Fig. 6b) at 2-10PDs, all exhibited increased nuclear oxidative DNA damage, even in untreated controls (arrowed, i-iv and v-viii). High proliferative DPSC subpopulation, A1, only exhibited similar nuclear FITC staining profiles to low proliferative DPSC subpopulations at 45-60PDs, both in untreated and H 2 O 2treated cultures (arrowed, Fig. 6c, i-iv and v-viii).

Catalase gene expression and activities in DPSCs
Catalase gene expression was maintained at relatively low levels in all high and low proliferative DPSCs analyzed (Fig. 10a). Although negligible basal catalase expression was determined in high proliferative DPSCs at 10-25PDs without H 2 O 2 treatment, untreated low proliferative DPSCs at 2-10PDs exhibited higher expression   Although catalase activities were at similarly low levels in high and low proliferative DPSCs at 10-25PDs and 2-10PDs without H 2 O 2 treatment (p > 0.05), significantly increased catalase activities were identified in high proliferative DPSCs at 45-60PDs without H 2 O 2 treatment (both p < 0.001, Fig. 10b). High proliferative DPSCs at 10-25PDs only demonstrated significantly increased total catalase activities with 50 μM H 2 O 2 (p < 0.01), versus untreated controls. However, equivalent catalase activities were shown between untreated and H 2 O 2 -treated high proliferative DPSCs at 45-60PDs (all p > 0.05 versus untreated controls), which were significantly higher than at early PDs with 50 μM and 200 μM H 2 O 2 treatments (p < 0.05 and p < 0.001, respectively). Low proliferative DPSCs at 2-10PDs without H 2 O 2 treatment exhibited equivalent catalase activities to untreated high proliferative DPSCs at 10-25PDs (p > 0.05), but were unable to induce further catalase activities with H 2 O 2 treatment (p > 0.05 versus untreated low proliferative DPSCs). Therefore, catalase activities for H 2 O 2 -treated low proliferative DPSCs at 2-10PDs were significantly lower than high proliferative DPSCs at 40-65PDs, irrespective of H 2 O 2 treatment (p < 0.001-0.05).

Discussion
Although DPSC susceptibility to replicative and oxidative stress-induced premature senescence has previously been recognized [12,16,[42][43][44], this is the first study to demonstrate the existence of inherent differences in oxidative stress responses and differential enzymic antioxidant profiles between DPSC sub-populations with contrasting proliferative capabilities, which subsequently impact on their respective multi-potency, stemness, and other cellular characteristics [12,13]. Despite the concept of DPSC proliferative and differentiation heterogeneity within dental pulp tissues being well-established [4][5][6], only recently have major variations in the proliferative potentials and susceptibilities to replicative senescence been confirmed between DPSC sub-populations, correlating with contrasting telomere lengths and the differentiation capabilities of individual populations [12]. Thus, it has been proposed that such high proliferative/ multi-potent DPSCs are responsible for the extensive expansion potential of heterogeneous populations (> 120PDs) in vitro [4][5][6], as less proliferative, uni-potent DPSCs would be selectively lost during extended culture [12,39]. However, as hTERT is unlikely to have a prominent role in maintaining telomere integrity in DPSCs [12,[14][15][16], we hypothesized that superior antioxidant capabilities contributed to the proliferative and multipotency capabilities of high proliferative DPSC subpopulations.
As with our previous study confirming variations in replicative senescence susceptibilities between high (> 80PDs) and low (< 40PDs) proliferative DPSCs [12], present findings identified similar variations in the relative susceptibilities of DPSC sub-populations to oxidative stress-induced premature senescence. Although all DPSC sub-populations exhibited accelerated susceptibilities to premature senescence in a H 2 O 2 dose-dependent manner, high proliferative DPSCs showed most resistance to H 2 O 2 -induced senescence, achieving 50-76PDs similar to untreated controls (> 80PDs). In contrast, low proliferative sub-populations collectively displayed accelerated premature senescence (4-32PDs with 50-200 μM H 2 O 2 ), even in untreated controls (only reaching . In support of their enhanced resistance to premature senescence, high proliferative DPSCs were further shown to possess fewer SA-β-galactosidase-positive cells and lacked the expression of p53 and p16 INK4a , at PDs where low proliferative DPSCs demonstrated increased detection with p21 waf1 , particularly following H 2 O 2 treatment [8]. MSC senescence is driven by tumor suppressors, such as p53, which promotes growth arrest by inducing p21 waf1 expression, inhibiting G 1 -S phase progression. Therefore, p53 and p21 waf1 regulate MSC expansion in an undifferentiated state. MSC senescence can also initiate p16 INK4a checkpoints, inducing senescence. Consequently, both p53 and p16 INK4a are regarded as the principal mediators of MSC senescence [16,42,43,45]. As p21 waf1 also maintains stem cell renewal [46,47], this may explain the presence of early p21 waf1 expression in all DPSCs analyzed. Nonetheless, contrasting p53 and p16 INK4a expression in high and low proliferative DPSC sub-populations further confirmed the early onset of premature senescence in low proliferative DPSCs. In agreement with previous reports, hTERT expression was undetectable in all DPSC subpopulations assessed [13][14][15][16].
A key reason identified to be responsible for contrasting proliferative responses and susceptibilities to replicative senescence were the mean telomere lengths between high and low proliferative DPSC sub-populations, with the superior telomere characteristics of high proliferative DPSCs permitting extended culture and protection from senescence [12]. In line with premature senescence occurring irrespective of extensive telomere shortening [8,10,17], all DPSCs largely retained their telomere length profiles during culture. Intriguingly, telomere lengths for high proliferative DPSC sub-population, A1, were the most influenced by extended culture and H 2 O 2 treatment, implying that this sub-population also underwent a degree of telomere-dependent senescence during culture. Alternatively, prolonged culture in H 2 O 2 can promote telomere shortening via oxidative damage and single-strand breaks [48,49]. Although we can only speculate on the extent to which telomere-dependent/ telomere-independent mechanisms contributed to telomere erosion and senescence in high proliferative subpopulation, A1, it may be assumed that both mechanisms are involved.
Further studies assessed the impact of premature senescence on the expression of stem cell markers in high and low proliferative DPSC sub-populations. In line with previous findings, all DPSCs were positive for MSC markers, CD73, CD90, and CD105, and negative for hematopoietic stem cell marker, CD45 [12,39]. Expression of MSC multi-potency markers, CD29, CD146, and CD271, were only evident in low proliferative DPSCs, as was the expression of stem cell differentiation regulator, CD166 [7,12,50,51]. However, all DPSC subpopulations showed strong positive gene expression for self-renewal/multi-potency marker, BMI-1 [16]. In terms of embryonic/neural crest markers, Oct4 was absent in high proliferative DPSCs, but expressed in all low proliferative DPSC sub-populations. In contrast, SSEA4 and Slug were positively expressed in high proliferative DPSCs and most low proliferative DPSCs. Oct4 and SSEA4 maintain embryonic self-renewal and pluripotency [52,53], while Oct4 and Slug are also implicated in promoting mesenchymal lineage commitment [53,54]. In agreement with previous reports of declined stem cell marker expression, stemness and multi-potency characteristics in MSC populations during senescence [12,16,39,44,51], increasing H 2 O 2 treatment and culture expansion reduced expression of CD73, CD105, SSEA4, and Slug in high proliferative DPSCs and Oct4 and CD271 in low proliferative DPSCs, potentially impacting on their stem cell and differentiation properties overall.
Having confirmed significant variations in DPSC subpopulation susceptibility to premature senescence, high proliferative DPSCs were further shown to exhibit resistance to oxidative stress-induced biomolecular damage that gradually diminished with culture expansion. In contrast, low proliferative DPSC sub-populations showed much earlier oxidative stress biomarker detection, even without H 2 O 2 treatment. Similar conclusions of elevated oxidative DNA and protein damage in low proliferative DPSC subpopulations have been reported by Raman Spectroscopy analysis [55]. Oxidative DNA damage is well-established to accompany cellular senescence [21,56], which could contribute to the early-onset of p53, p21 waf1 , and p16 INK4a induction and increased premature senescence in low proliferative DPSC sub-populations [27][28][29][30][31]36]. Oxidative protein damage, as particularly evident in low proliferative DPSCs, is also a well-documented occurrence during cellular senescence, due to oxidized protein modification and accumulation [22,57].
The relative susceptibilities of DPSC sub-populations to oxidative damage and premature senescence suggested that such responses were related to contrasting antioxidant defense mechanisms between high and low proliferative DPSCs. High proliferative DPSCs were demonstrated to possess superior abilities to induce certain enzymic antioxidant expression and activities, compared to low proliferative DPSCs. The ability to upregulate antioxidant expression to counteract ROS is a fundamental concept of oxidative stress, including resistance to cellular senescence [27][28][29][30][31][32][33][34][35][36][37]. SOD profiles demonstrated distinct differences between high and low proliferative DPSCs, with low SOD1 and SOD3 levels particularly detectable in low proliferative DPSCs. In contrast, only high proliferative DPSCs at 10-25PDs demonstrated significantly induced SOD2 expression (10-15-fold) with H 2 O 2 treatment. Such findings imply that SODs predominantly localized within cytosolic (SOD1) and extracellular (SOD3) regions do not contribute to the antioxidant status of high proliferative DPSCs, although induction of SOD1 and SOD3 expression in untreated and H 2 O 2 -treated low proliferative DPSCs imply that these sub-populations are experiencing oxidative stress [55]. The relatively high SOD2 induction in high proliferative DPSCs strongly suggests that SOD2 is a prominent mediator of antioxidant activity within these sub-populations [58]. As SOD1 and SOD2 are ubiquitously expressed by aerobic cells, the low SOD1 levels in high proliferative DPSCs is intriguing, although the absence of SOD3 can be explained by its more specific cellular expression profiles [19,24]. Due to the limited SOD1 and SOD3 detection, it is likely that most SOD activity induced in H 2 O 2 -treated high proliferative DPSCs is accountable by upregulated SOD2 expression. In contrast, expression and protein analyses suggest that SOD1 is a principal contributor to SOD activities in low proliferative DPSCs.
Catalase profiles demonstrated higher expression and activities in low proliferative DPSCs, although only relatively low levels of catalase were detectable in DPSCs overall, even with H 2 O 2 treatment. Similar catalase expression/activity profiles have been reported in high and low proliferative bone marrow-derived MSCs [58]. However, despite being a potent cytosolic H 2 O 2 detoxifying antioxidant, catalase is particularly susceptible to downregulation and inactivation by ROS [59,60], which may be responsible for the low catalase expression/activity levels detected. Thus, although catalase appears to have a relatively minor role in mediating antioxidant responses in high proliferative DPSCs, as with SOD1, induction of limited catalase expression in untreated and H 2 O 2 -treated low proliferative DPSCs may imply that these sub-populations are already experiencing elevated oxidative stress [55].
Analysis of glutathione-metabolizing enzymes demonstrated that GPX, GSR, and GSS expression and GPX activities were undetectable in high proliferative DPSCs. Similarly, although low proliferative DPSCs exhibited GPX1, GPX3, GPX4, GSR, and GSS expression and GPX4, GSR, and GSS induction with increasing H 2 O 2 treatment, gene expression and GPX activities were relatively low overall. Such findings imply that glutathionerelated enzymes are not major contributors to the antioxidant status of high proliferative DPSCs, although low proliferative DPSCs may be more reliant on these antioxidant mechanisms. However, only high proliferative DPSCs at 10-25PDs significantly induced the expression of GSTZ1 with H 2 O 2 treatment (100-125-fold). While GPXs reduce glutathione (GSH) and GSR exerts antioxidant defenses through the decomposition of H 2 O 2 and hydroperoxides [25,26], GSTZs primarily detoxify xenobiotics and endobiotics within the cytosol and mitochondria [61].
Mitochondria are established as the principle cellular source of ROS during senescence [58,62,63]. Thus, mitochondrial-specific SOD2 is acknowledged as the primary enzymic antioxidant against oxidative damage within mitochondria and the prevention of cellular senescence [19,24,32,44,58]. Increased GSTZ1 expression, especially mitochondrial GSTZ1, has also been strongly associated with decreased human aging and prolonged longevity, partly due to reduced telomere shortening [64,65]. Furthermore, GSTZ1 −/− mice possess alterations in mitochondrial ultrastructure, size, and activity, confirming the protective roles of GSTZ1 in mitochondria [61,66]. Therefore, our findings imply that mitochondrial-derived ROS are significant mediators of oxidative damage and premature senescence in low proliferative DPSCs, whereas high proliferative DPSCs are more resistant due to significant adaptations in SOD2 and GSTZ1 expression, leading to the extended maintenance of proliferative, stem cell, multipotency, and other cellular characteristics [12,13]. However, the absence of SOD2 and GSTZ1 inductions with prolonged culture expansion suggest that these adaptive antioxidant mechanisms become defective, leading to increased susceptibility to oxidative damage and premature senescence. Similar findings have been reported in MSCs from other sources, with senescent cells exhibiting lower SOD, catalase, and GPX expression, resulting in reduced antioxidant status and overall increases in oxidative stress [34][35][36][37].
Despite the findings presented herein, a limitation of the present study is that it has compared oxidative stress-induced biomolecular damage and SOD2/GSTZ1 profiles within high proliferative DPSCs derived from only one patient (patient A). As high proliferative/multipotent DPSCs are regarded as minority sub-populations within dental pulp tissues [7,12,55], current screening protocols are not completely efficient for the guaranteed isolation of high proliferative/multi-potent DPSC subpopulations from the dental pulp tissues of all patient donor teeth [67]. Consequently, low proliferative/uni-potent DPSCs are usually the predominant subpopulations isolated and as a result, high and low proliferative DPSC sub-populations were not compared from all collected patient teeth. Thus, the true nature of such high proliferative/multi-potent minority DPSC subpopulations advocates more detailed investigations to confirm their reproducible isolation, presence, and regenerative characteristics across a wider number of patient-matched high and low proliferative DPSCs from the same donor teeth, in order to fully establish the relationship between oxidative damage, SOD2 and GSTZ1 profiles and how these impact on the overall PD capabilities and multi-potent differentiation capabilities of individual DPSC sub-populations. Indeed, we can only speculate on the underlying reasons for such differences between high proliferative/multi-potent and low proliferative/uni-potent DPSC sub-populations at present, as intrinsic features, such as those associated with patient donor characteristics and/or their developmental origins and stem cell niche sources within dental pulp tissues could all be influential factors and warrant additional consideration [4,5,7,12,67]. glutathione; GSR: Glutathione reductase; GSS: Glutathione synthetase; GST: Glutathione S-transferase; GSTZ1: Glutathione S-transferase ζ1; hTERT: Reverse transcriptase human telomerase catalytic subunit; αMEM: αmodified Minimum Essential Medium; MSC: Mesenchymal stem cell; PBS: Phosphate-buffered saline; PDs: Population doublings; pRb: Retinoblastoma protein; qRT-PCR: Quantitative real-time polymerase chain reaction; ROS: Reactive oxygen species; RQ: Relative fold changes in gene expression; RT-PCR: Reverse transcription polymerase chain reaction; SA-β-galactosidase: Senescence-associated-β-galactosidase; SASP: Senescence-associated secretory phenotype; SDS-PAGE: Sodium dodecyl sulfate-polyacrylamide gel electrophoresis; SEM: Standard error of mean; SOD: Superoxide dismutase; TBS: Tris-buffered saline; TRF: Telomere restriction fragment; TRITC: Tetramethylrhodamine