Low Oxidative Stress-Mediated Proliferation Via JNK-FOXO3a-Catalase Signaling in Transplanted Adult Stem Cells Promotes Wound Tissue Regeneration

Low oxidative stress promotes proliferation of BMSCs and HSPCs

To evaluate the growth pattern of BMSCs and HSPCs in the presence of H₂O₂, the cells were treated for a maximum period of 96 h as previously described (11). Cells were harvested at a regular interval, and cell count was determined. Both BMSCs (Fig. 1A) and HSPCs (Fig. 1B) depicted a noticeable increase in cell growth, with a reduction in doubling time to 48 h when treated with 10 μM H₂O₂ as compared with their respective controls (≥96 h). At higher concentrations of H₂O₂ (100 μM), growth rates of both BMSCs (Fig. 1A) and HSPCs (Fig. 1B) were markedly decreased. Vit. C or DHA at an equimolar concentration could revert the low H₂O₂-induced increased growth rate of both BMSCs (Fig. 1A) and HSPCs (Fig. 1B). Further to corroborate the growth pattern of BMSC and HSPC, CFU and bromodeoxyuridine (BrdU) cell proliferation assays were performed. As observed in growth curve kinetics of 10 μM but not 100 μM concentration of H₂O₂, a significant increase in CFU and BrdU incorporation was depicted in both BMSCs (Fig. 1C, E) and HSPCs (Fig. 1D, 1). Equimolar concentration (10 μM) of Vit. C or DHA could rescue the effect of H₂O₂. These results clearly suggest that the specificity of H₂O₂-mediated oxidative stress promotes the proliferative potential of BMSCs and HSPCs that could be perturbed by the exogenous supplementation of antioxidants.

OPEN IN VIEWER

FIG. 1.

Low concentration of H₂O₂ induces proliferation of BMSCs and HSPCs. Significant increase in cell growth was observed in both bone marrow-derived BMSCs (A) and HSPCs (B) as depicted by growth curve analysis. Antioxidant effects of Vit. C or DHA could significantly reduce the H₂O₂-induced growth rate of BMSCs and HSPCs at 10 μM concentration. Similar increase in H₂O₂-induced colony formation units was observed in BMSCs (C) and HSPCs (D) that was perturbed by exogenous Vit. C or DHA in colony formation assay. Ten micromolar H₂O₂-induced cell proliferation was also observed in BMSCs (E) and HSPCs (F) that was reversed by Vit. C and DHA in BrdU cell proliferation assay. Results are depicted as percent proliferation of control (n = 4–6 replicates per isolation; p < 0.05 as compared with *control and ^#10 μM H₂O₂-treated group; p < 0.01 as compared with **control). BMSC, bone marrow stromal stem/progenitor cells; BrdU, bromodeoxyuridine; DHA, dehydroascorbic acid; H₂O₂, hydrogen peroxide; HSPC, hematopoietic stem and progenitor cells; Vit. C, vitamin C. To see this illustration in color, the reader is referred to the web version of this article at www.liebertpub.com/ars

Low oxidative stress promotes cell cycle progression in BMSCs and HSPCs

The earlier observation of 10 μM H₂O₂-induced proliferation incited us to explore the cell cycle status of these sorted bone marrow cell populations. Ten micromolar H₂O₂ significantly increased cell populations in S and G₂/M phases of the cell cycle that was perturbed in the presence of Vit. C (Supplementary Fig. S1A; Supplementary Data are available online at www.liebertpub.com/ars). To substantiate the cell cycle status, we investigated the changes in cell cycle gene expression in BMSCs and HSPCs. Expression of cyclins B1, D2, and E1 was increased by several folds with 10 μM H₂O₂ that was attenuated by an equimolar concentration of Vit. C (Supplementary Fig. S1B). In addition, 10 μM H₂O₂-dependent decrease in expression of p16 and p27 cyclin-dependent kinase inhibitors (CKI) and an increase in proliferating cell nuclear antigen (PCNA) expression were observed, indicating the cell passage through S-G₂/M phases of the cell cycle (Supplementary Fig. S1C). The overall increase in cyclins, PCNA, and decrease in CKI expression by 10 μM H₂O₂ promote the bone marrow-derived BMSCs and HSPCs to undergo cell expansion, which otherwise remains in the G₀/G₁ phase of the cell cycle. The mechanism of this cellular effect by oxidative stress was further supported by the ROS quencher-induced reversal of specific cell cycle gene induction.

Cellular redox status in H₂O₂-treated BMSCs and HSPCs

To evaluate the intracellular and extracellular ROS levels along with glutathione (GSH) content in sorted BMSCs and HSPCs exposed to H₂O₂, DCFH-DA staining, amplex red assay, and glutathione assay were performed. Significantly high concentration-dependent intracellular ROS levels were observed after 24 and 48 h post-treatment of H₂O₂ in BMSCs (Supplementary Fig. S2A) and HSPCs (Supplementary Fig. S2B) that was attenuated by Vit. C or DHA. The extracellular ROS level (H₂O₂ concentration) depicted a concentration-dependent significant increase that was overturned by addition of Vit. C or DHA as compared with control at 24 h. Further at 48 h post-treatment, the extracellular ROS level decreased and was comparable to the control group, indicating the utilization of H₂O₂ by BMSCs (Supplementary Fig. S2C) and HSPCs (Supplementary Fig. S2D). In line with the H₂O₂-induced increased ROS production, a concentration-dependent significant decrease of cellular GSH content was restored in BMSCs (Supplementary Fig. S2E) and HSPCs (Supplementary Fig. S2F) by addition of Vit. C or DHA as compared with the control at 48 h post-treatment. These data suggest that even 10 μM H₂O₂ causes significant oxidant perturbation in BMSCs and HSPCs that can, subsequently, activate downstream signal transduction pathways.

Low oxidative stress-induced activation of molecular signal transduction pathways in BMSCs and HSPCs

We evaluated the role of oxidative stress on growth-promoting and proliferative signal transduction pathways in BMSCs and HSPCs using immunoblot analysis. Ten micromolar H₂O₂ treatment in these cells led to an increased phosphorylation of AKT and JNK (Fig. 2A). Further, a sub-lethal or high concentration of H₂O₂ (100 μM) could phosphorylate JNK but not AKT, indicating a selective activation of JNK stress kinase (Fig. 2A), which was further confirmed by densitometric analysis (BMSCs-Supplementary Fig. S3A and HSPCs-Supplementary Fig. S3B). In addition, activation of ERK1/2 and p38 MAPK under similar conditions yielded insignificant change of these signaling mediators in both BMSCs and HSPCs. To confirm whether the earlier observation holds true, we treated HSPCs and BMSCs with H₂O₂ in both the presence and absence of inhibitors such as Wortmannin (AKT), SP600125 (JNK), PD98059 (ERK1/2) (48), and SB203580 (p38 MAPK) (12). Indeed, wortmannin and SP600125 at 100 nM and 30 μM, respectively, could significantly decrease the H₂O₂-induced proliferation in BMSCs (Fig. 2B) as well as HSPCs (Fig. 2C) as observed in both growth curve analysis and BrdU cell proliferation assays (Fig. 2D), suggesting pathway specificity. These observations indicate that 10 μM H₂O₂-induced proliferation is mediated by concerted action of both AKT and JNK signaling pathways.

OPEN IN VIEWER

FIG. 2.

Role of AKT and JNK activation on H₂O₂-induced BMSC and HSPCs proliferation. (A) Expression levels of phosphorylated and total protein levels of AKT, ERK1/2, p38 MAPK, and JNK were determined by Western blot analysis. Ten micromolar H₂O₂-induced AKT and JNK activation (phosphorylation) that was reverted by 10 μM Vit. C in both BMSCs and HSPCs. Growth curve analysis of cells exposed to10 μM H₂O₂ along with pathway inhibitors in BMSCs (B) and HSPCs (C) depicted a decreased growth rate in the presence of wortmannin (AKT inhibitor) and SP600125 (JNK inhibitor) treatment. PD98059 (ERK1/2 inhibitor) and SB23058 (p38 MAPK inhibitor) did not reduce the 10 μM H₂O₂-induced growth rate of both BMSCs and HSPCs. (D) BrdU cell proliferation assay depicted a significant decrease in BrdU incorporation in both BMSCs and HSPCs exposed to 10 μM H₂O₂ along with wortmannin and SP600125, suggesting a pathway specificity (n = 4-6 replicates per isolation; p < 0.05 as compared with *control and ^#10 μM H₂O₂-treated group). JNK, c-Jun N terminal kinase. To see this illustration in color, the reader is referred to the web version of this article at www.liebertpub.com/ars

Cellular antioxidants facilitate BMSC and HSPC proliferation at low oxidative stress

Further, we investigated the status of the redox genes that are essential to detoxify the H₂O₂ as well as its effect on the downstream cell signaling. Among the different isoforms of mammalian SODs, an H₂O₂-mediated concentration-dependent increase in the expression of SOD2, the mitochondrial isoform was observed in both BMSC and HSPC populations, which were decreased by an equimolar concentration of Vit. C (Supplementary Fig. S4A). Similarly, the expression of GPx2 (mitochondrial isoform of GPx) depicted a 2.9-fold high in both BMSCs and HSPCs as compared with the control in the presence of H₂O₂, which was not abrogated by an equimolar concentration of Vit. C (Supplementary Fig. S4B). Expression of catalase was 3.75-fold higher in BMSCs and 3.9-fold higher in HSPCs than the respective control at 10 μM exposure of H₂O₂, which was mitigated in the presence of an equimolar concentration of Vit. C, indicating that low oxidative stress induces its expression (Fig. 3A). However, at sub-lethal H₂O₂ concentration, catalase expression was decreased in both BMSCs and HSPCs, which substantiates with the existing literature that H₂O₂ at a high concentration downregulates catalase expression by epigenetic regulation via methylation of CpG islands (35).

OPEN IN VIEWER

FIG. 3.

Low concentration of H₂O₂ potentiates the expression of antioxidant catalase, which when inhibited reversed H₂O₂-induced proliferative effect on BMSCs and HSCs. 10 μM concentration of H₂O₂ induced catalase mRNA (A) and protein (B) expression, whereas a high concentration did not. (C) Biochemical enzyme activity of catalase was evaluated in BMSCs and HSPCs treated with H₂O₂, Vit. C, and 3-AT. Catalase activity was significantly high in cells treated with 10 μM H₂O₂. Vit. C at an equimolar low concentration (10 μM) and 3-AT, a catalase inhibitor (30 μM), significantly reversed the H₂O₂-induced catalase activity in these cell populations. A further increase to 100 μM H₂O₂ decreased the catalase activity. (D) Similarly, a decrease in H₂O₂-induced colony-forming efficiency was observed in the presence of 3-AT. Next, H₂O₂-induced growth pattern of BMSCs (E) and HSPCs (F) in the presence of 3-AT depicted a decrease in growth rate. Further addition of PEG-catalase reversed the inhibition shown by 3-AT treatment. (G) Addition of 3-AT attenuated low H₂O₂-induced BrdU cellular incorporation in BMSCs and HSPCs. Addition of PEG catalase reversed the inhibitory effect imparted by 3-AT. Further, siRNA-CAT (siRNA against catalase gene) transfected BMSCs and HSPCs depicted a significant decrease in BrdU incorporation (n = 3–6 replicates per isolation; p < 0.05 as compared with *control and ^#10 μM H₂O₂-treated group). 3-AT, 3 Amino-triazole. To see this illustration in color, the reader is referred to the web version of this article at www.liebertpub.com/ars

Further, in parallel with their transcript levels, we also evaluated the activity of these antioxidant enzymes SOD, GPx, and catalase in cell extracts. Although a concentration-dependent increase in activity of SOD was observed in HSPCs and BMSCs treated with H₂O₂, an equimolar concentration of Vit. C could not reverse the H₂O₂-induced increase in total SOD activity (Supplementary Fig. S4C). Similarly, total GPx activity depicted an insignificant change in BMSCs (Supplementary Fig. S4D) but in HSPCs a significant increase was observed only at a high concentration (100 μM) of H₂O₂ (Supplementary Fig. S4D). Increased expression of catalase translated equally with increased protein expression (Fig. 3B) and increased biochemical activity at 10 μM H₂O₂ that was reversed by exogenous antioxidants in BMSCs and HSPCs (Fig. 3C). These observations clearly indicate the role of the antioxidant enzyme, catalase in regulating the oxidative stress caused due to 10 μM H₂O₂ in BMSCs and HSPCs, thereby modulating cellular proliferation.

Catalase inhibition sensitizes BMSCs and HSPCs to low oxidative stress

To further investigate the role of catalase in low oxidative stress-induced cell proliferation, a pharmacological inhibitor of catalase, 3-Aminotriazole (3-AT) was used (41). A low concentration of H₂O₂-stimulated cell proliferation was significantly decreased by 3-AT (30 μM), as depicted by a decrease in the percent cell population in S-G₂/M phase of the cell cycle (Supplementary Fig. S5A). At a biochemical level, catalase enzyme activity was also observed to be significantly decreased in BMSCs and HSPCs treated with 10 μM H₂O₂ and 3-AT (Fig. 3C). Next, colony formation (Fig. 3D) and growth curve assays also depicted a decrease in cell number when treated with 10 μM H₂O₂ in combination with 30 μM 3-AT in both BMSC (Fig. 3E) and HSPC (Fig. 3F). A similar outcome was observed in BrdU cell proliferation assay (Fig. 3G), although 3-AT treatment alone did not depict any significant difference in cell proliferation (Supplementary Fig. S5B). The pharmacologic administration of PEG catalase rescued the inhibitory effect on cell proliferation imparted by 3-AT (Fig. 3E–G). Catalase siRNA transfection of both BMSCs and HSPCs resulting in gene silencing (Supplementary Fig. S5C) led to a decrease in 10 μM H₂O₂-induced cell proliferation (Fig. 3G). These results clearly suggest that an increased cell cycle progression and expansion of BMSCs and HSPCs in the presence of 10 μM H₂O₂ but not at higher 100 μM H₂O₂ is correlated with increased catalase expression and activity.

To determine the effects of catalase inhibition on proliferative signaling mechanisms of AKT and JNK activation, we treated both BMSCs and HSPCs with 3-AT in the presence/absence of 10 μM or 100 μM concentration of H₂O₂. Catalase inhibition in these cell types depicted a downregulation of the AKT but not JNK signaling pathway (Fig. 4A), as evident from the densitometric analysis (Supplementary Fig. S5D). This observation suggests that in the absence of active catalase, even 10 μM concentration of H₂O₂ sensitizes the cells by inducing JNK stress kinases alike a sub-lethal (100 μM) concentration of H₂O₂ (Supplementary Fig. S5D). Similarly, decreased AKT activation whereas increased JNK activation was observed with silencing of catalase using CAT-siRNA at molecular level (Fig. 4B) and densitometric analysis or repeated blots (Supplementary Fig. S5E). To confirm AKT activation only at 10 μM H₂O₂ in both BMSC and HSPC with active catalase, we evaluated the AKT kinase activity in the presence/absence of 3-AT (10). Significantly high AKT kinase activity was observed in the presence of 10 μM H₂O₂ that was inhibited by 3-AT or wortmannin (positive control) (Fig. 4C). Interestingly, JNK inhibition also decreased AKT activation, indicating an existence of signaling crosstalk (Fig. 4C). Further, at molecular level, JNK (23) and AKT (47) dominant negative (DN) mutants could also decrease the 10 μM H₂O₂-induced AKT activity (Supplementary Fig. S6A). Next, we evaluated the levels of intracellular ROS in the presence of catalase inhibition. Significantly similar high intracellular ROS levels were observed in both BMSC (Supplementary Fig. S6B) and HSPC (Supplementary Fig. S6C) treated with 10 μM or 100 μM H₂O₂ along with 3-AT administration, suggesting an increased ROS-mediated AKT inactivation. Ten micromolar H₂O₂-induced increased AKT activation (phosphorylation) was perturbed by molecular inhibition using AKT and JNK DN mutants in both BMSC (Supplementary Fig. S6D) and HSPC (Supplementary Fig. S6E) along with cell proliferation (Supplementary Fig. S6F). Interestingly, molecular inhibition of JNK also led to decreased cell proliferation, indicating its involvement via catalase downregulation-mediated effect.

OPEN IN VIEWER

FIG. 4.

Low concentration of H₂O₂ led to an increased JNK activation and FOXO3a nuclear translocation. (A) Western blot analysis revealed that 10 μM H₂O₂-induced AKT activation was reverted by addition of 3-AT. On the contrary, catalase inhibition led to insignificant change in JNK phosphorylation. (B) Silencing the catalase at molecular level by using CAT siRNA transfection depicted similar results as 3-AT. (C) 10 μM H₂O₂-induced AKT kinase activity was inhibited in the presence of inhibition of catalase (3-AT) or JNK (SP600125) or positive control AKT (wortmannin) in both BMSCs and HSPCs. (D) A dose-dependent increase in H₂O₂-mediated FOXO3a nuclear translocation was abrogated in the presence of Vit. C that sequestered FOXO3a in the cytoplasm. Pharmacological (SP600125) as well as molecular (pCMV-DN JNK) inhibition of JNK prevented FOXO3a nuclear translocation, whereas AKT inhibition using wortmannin or pCMV-AktDN retained FOXO3a into the nucleus. Cells were stained with anti-FOXO3a rabbit polyclonal antibody followed by addition of AlexaFluor 488-labeled anti-rabbit secondary antibody. Nuclei were stained with DAPI. Quantification of nuclear localization along with Western blot images was evaluated by using NIH Image J software. CAT, catalase; DAPI, 4′,6-diamidino-2-phenylindole; FOXO3a, forkhead box O3a. To see this illustration in color, the reader is referred to the web version of this article at www.liebertpub.com/ars

FOXO3a nuclear translocation induces transcriptional upregulation of catalase at low oxidative stress

In the JNK stress signaling cascade, activated JNK is known to translocate FOXO3a, a transcription factor to the nucleus for transcriptional upregulation of FOXO3a target genes (25). Therefore, we determined the concentration-dependent effect of H₂O₂-induced activation of JNK on FOXO3a nuclear translocation. As expected, 10 μM and 100 μM H₂O₂ depicted a concentration-induced nuclear translocation of FOXO3a (Fig. 4D and Supplementary Fig. S7A–C) whereas Vit. C led to FOXO3a retention in the cytoplasm (Fig. 4D, Supplementary Fig. S7A). Pharmacological (Fig. 4D and Supplementary Fig. S7A) as well as molecular (Fig. 4D and Supplementary Fig. S7A–C) inhibition of JNK inhibited the nuclear translocation of FOXO3a, whereas a similar inhibition of AKT did not have any effect on FOXO3a nuclear translocation (Fig. 4D and Supplementary Fig. S7A–C). Although activation of AKT is known to impart an inhibitory effect on FOXO3a nuclear translocation, recent studies suggest that a simultaneous JNK activation overrides AKT signaling, thereby dominating the FOXO3a nuclear translocation (4). Our results too clearly suggest the mechanism of H₂O₂-induced oxidative stress-mediated JNK activation, which, in turn, leads to nuclear translocation of FOXO3a that acts as a transcriptional co-regulator of its target gene expression. In our in silico analysis (data not shown) as well as other studies, an FOXO3a binding site is suggested in the promoter region of catalase gene (39). To further confirm this, we evaluated the catalase promoter that contained three potential FOXO3a binding sites: one at a distal region, distal binding site (DBS) and two others at a proximal region, proximal binding site (PBS1 and PBS2) (Fig. 5A). Chromatin immunoprecipitation (ChIP) analysis revealed that at 10 μM H₂O₂ concentration, FOXO3a binds with a relatively higher affinity to catalase promoter at DBS in HSPCs as compared with BMSCs (Fig. 5B). However, a differential binding of FOXO3a to the PBS1 and PBS2 on catalase promoter was observed at 10 μM H₂O₂ concentration. A high binding of FOXO3a to PBS2 in BMSCs (Fig. 5C) and PBS1 in HSPCs (Fig. 5D) was observed. Increased H₂O₂ concentration consequently led to decreased FOXO3a binding to catalase promoter (Fig. 5B–D). Vit. C and JNK inhibitor, SP600125 in the presence of 10 μM H₂O₂ concentration, decreased FOXO3a binding to catalase promoter, indicating the 10 μM H₂O₂-mediated transcriptional upregulation of catalase via JNK-FOXO3a signaling. This was further confirmed by a decreased catalase gene expression in the presence of JNK inhibitor (SP600125) but not AKT inhibitor (wortmannin) in 10 μM H₂O₂-treated BMSC and HSPCs (Fig. 5E), suggesting that although 10 μM H₂O₂-induced activation of JNK and AKT was observed in these proliferating cells, only JNK but not AKT imparts a direct regulation on catalase expression. To further substantiate this observation, we evaluated catalase mRNA and protein expression in the presence of JNK and AKT DN mutants. Molecular inhibition of JNK but not AKT led to a decrease in catalase mRNA (Supplementary Fig. S7D) and protein (Supplementary Fig. S7E) expression in 10 μM H₂O₂-treated BMSC and HSPC, which corroborated further with catalase activity (Supplementary Fig. S7F). These results indicate that physiological relevance of AKT activation in 10 μM H₂O₂-treated BMSC and HSPC might have a direct regulation on cell proliferation.

OPEN IN VIEWER

FIG. 5.

Low concentration of H₂O₂ potentiates FOXO3a binding to catalase promoter. (A) Schematic representation of catalase promoter depicting different FOXO3a binding sites upstream to transcription start site. ChIP analysis depicted an increased binding of FOXO3a on the catalase promoter at the (B) distal binding site (DBS, −1980 to −1974), (C) proximal binding site 2 (PBS2, −798 to −792), and (D) PBS1 (−612 to −606). (E) Low H₂O₂-induced catalase mRNA expression was abrogated in the presence of JNK inhibitor, SP600125 but not AKT inhibitor, wortmannin. (F) AKT as well as catalase inhibition significantly decreased the 10 μM H₂O₂-induced cyclins B1, D2, and E2 expression in both BMSCs and HSPCs. (G) 10 μM concentration of H₂O₂-mediated decreased expression of CKIs, p16, and p27 was markedly reversed by AKT or catalase inhibition (n = 3 replicates per isolation; p < 0.05 as compared with *control; ^#10 μM H₂O₂-treated group).

AKT and catalase inhibition downregulates cell proliferation-related genes at low oxidative stress

AKT inhibition-mediated decreased cell proliferation in 10 μM H₂O₂-treated BMSC and HSPC was confirmed by the observation of decreased expression of cyclins B1, D2, and E2 (Fig. 5F) and PCNA (Fig. 5G) along with concomitant increased expression of CKIs p16 and p27 (Fig. 5G). Molecular inhibition of AKT also led to a decreased expression of cyclins (Supplementary Fig. S7G) and increased expression of CKIs (Supplementary Fig. S7H), thereby confirming the role of AKT in cell proliferation. Interestingly, JNK molecular inhibition indirectly led to a decrease in cyclin expression and an increase in CKIs expression via a decrease in catalase expression (Supplementary Fig. S7F, G). As observed earlier, downregulation of catalase led to inactivation of AKT; 3-AT treatment also resulted in decreased cyclins (Fig. 5F), PCNA (Fig. 5G) and increased CKIs expression (Fig. 5G), indicating a simultaneous induction of the JNK–FOXO3a–catalase signaling cascade along with AKT–cyclins–cell proliferation in the presence of 10 μM H₂O₂-induced low oxidative stress in BMSC and HSPC wherein perturbation of the former signaling regulates the latter.

Low oxidative stress preconditioned BMSCs and HSPCs enhance wound tissue repair

One of the major causes of poor endurance of stem cells in vivo in an injury microenvironment is the poor expansion of transplanted stem cells. Exposure to low oxidative stress in vitro (preconditioning with 10 μM H₂O₂) before transplantation at the wound site was hypothesized to increase the proliferation of BMSCs and HSPCs. Thus, we utilized a murine model of excisional splinting wound as described earlier (9) wherein wounds heal through granulation and re-epithelialization, a process similar to that which occurs in humans (43). Mice transplanted with preconditioned BMSCs and HSPCs showed an accelerated wound closure postsurgery day 7 as compared with mice transplanted with naive BMSCs and HSPCs only (Fig. 6A). Further temporal quantification of the regenerated wound area depicted a 75% wound closure in preconditioned BMSCs as compared with 56% in naive BMSCs at postsurgery day 7 (Fig. 6B). Similar results were observed with preconditioned HSPCs representing an increased wound closure of 71% at postsurgery day 7 (Fig. 6B). Histopathological analysis of regenerated wound tissue using hematoxylin and eosin staining (Fig. 6C, lower magnification images–Supplementary Fig. S8A) and the quantitation of images (Fig. 6D) clearly demonstrated an increased proliferation of cells in preconditioned BMSCs and HSPCs as compared with naive BMSCs and HSPCs and/or wound-control (nonstem cells transplanted) groups. Also, the preconditioned stem cells transplanted group demonstrated an increased collagen deposition as demonstrated by Masson's Trichrome staining (Fig. 6E) and its quantitation (Fig. 6F). Overall ROS generated at the wound site was evaluated by using dihydroxy ethidium (DHE) staining. Remarkably, preconditioned BMSCs and HSPCs transplanted groups depicted a marked reduction in DHE staining as compared with naive BMSCs or HSPCs transplanted groups and the wound control (Fig. 6G). Concurrently, a significant increase in the expression of antioxidant enzymes such as MnSOD (SOD2) (≥5.5–fold), GPx1 (∼8–fold), and GPx2 (∼3.8–fold) was observed in preconditioned HSPCs and BMSCs transplanted groups (Supplementary Fig. S8B). However, the gene expression levels did not correlate with the activity of these enzymes—SOD (Supplementary Fig. S8C) and GPx (Supplementary Fig. S8D). Interestingly, an increase in catalase gene expression in preconditioned BMSCs (10-fold) and/or HSPCs (13-fold) transplanted groups (Fig. 6H) substantiated well with catalase activity (31.9 and 31.5 μmoles/min/μg of protein, respectively) as compared with naive transplanted or wound-control group (Fig. 6I). Thus, preconditioning BMSCs or HSPCs with low concentrations of H₂O₂ induced the antioxidant enzyme, catalase. These cells when transplanted at the wound site alleviated the oxidative stress, thereby promoting wound tissue regeneration.

OPEN IN VIEWER

FIG. 6.

10 μ M H₂O₂ preconditioned BMSCs or HSPCs transplantation improves wound repair and antioxidant defense system. (A) A representative image of wound healing in C57BL/6J mouse at postsurgery days 0 (upper panel) and 7 (lower panel). (B) Quantitation of wound area at different time points postwound generation (Days 0, 3, and 7) was performed by using NIH image J software. The results were represented as percent wound closure. (C) Representative images of hematoxylin and eosin-stained section of the regenerated wound tissue at postsurgery day 7 in various groups. (D) Quantitative analysis using NIH image J software depicted a significant increase in the number of H&E positively stained cells in preconditioned BMSCs and HSPCs transplanted groups. (E) Representative images of Masson's trichome (MT) stained sections of the regenerated wound tissues depicting the collagen deposition. (F) Quantification of the number of MT positively stained cells in various groups. (G) A representative image of DHE-stained sections to evaluate the ROS levels at the wound site. (H) Relative mRNA expression level of antioxidant enzyme, catalase was evaluated from regenerated wound tissue by using qRT-PCR analysis. Interestingly, 10 μM H₂O₂ preconditioned HSPCs transplanted group depicted significantly high expression of catalase. (I) Catalase activity depicted a significant increase in naive as well as 10 μM H₂O₂ preconditioned BMSCs or HSPCs transplanted (Tx) groups as compared with the wound control at postsurgery day 7 (Wd-Cont - Wound Control, PC - Preconditioned; n = 3 replicates per wound, N = 5 mice/group; p < 0.05 as compared with *wound control; ^#BMSC or HSPC transplanted group; **p < 0.01 as compared with wound control). To see this illustration in color, the reader is referred to the web version of this article at www.liebertpub.com/ars

Low oxidative stress preconditioned BMSCs and HSPCs improve engraftment at the wound site

To determine the engraftment of transplanted stem cells at the wound site, BMSCs and/or HSPCs chimera was generated by using donor cells of male C57BL/6J mice transplanted onto recipient female C57BL/6J mice. DNA isolated from the wound tissue at 7 days postsurgery was subjected to quantification of Y-linked zinc finger protein (Zfy-1) gene and Bcl-2 (as reference control) by using real-time PCR. Preconditioned BMSCs and HSPCs transplanted groups depicted an increased engraftment of male donor cells as compared with naive BMSCs and HSPCs transplanted groups that were abrogated when these cells were also pretreated simultaneously with an exogenous antioxidant, Vit. C (Fig. 7A). Further, engraftment of transplanted GFP-expressing BMSCs and HSPCs co-expressing other stem cell markers as reported earlier (8) were also evaluated in the regenerated wound sections. At postsurgery day 7, an increased co-localization of GFP-CAT was observed in preconditioned BMSCs and HSPCs as compared with naive BMSCs and HSPCs transplanted groups, respectively, or the wound control (Fig. 7B and quantitation–Supplementary Fig. S8E), thereby suggesting an increased cellular antioxidant defense mechanism in the transplanted cells. In an effort to understand whether preconditioned BMSCs and HSPCs transplanted groups with a high expression of catalase could decrease the oxidative stress, we co-stained the tissue sections with DHE. Indeed, preconditioned BMSCs and HSPCs transplanted groups depicted a marked decrease in DHE staining (Fig. 7C) as compared with naive BMSCs and HSPCs transplanted groups or the wound control (Fig. 7D). Simultaneous supplementation of Vit. C to 10 μM concentration of H₂O₂-mediated preconditioning of BMSCs and HSPCs in vitro when transplanted markedly decreased the engraftment of cells as evident from less GFP-positive cells and increased DHE staining. This phenomenon corroborated well with the in vitro observation (Fig. 7C, D). Next, the “antioxidant” nature of Vit. C as an ROS quencher was evaluated for any beneficial effects during wound repair if administered parenterally at a pharmacological dose (200 mg/kg body wt, i.p., per day × 7 days). We observed a significant increase in wound closure at postsurgery day 7 as compared with the wound control (Supplementary Fig. S9A) and increased collagen deposition at the wound bed (Supplementary Fig. S9B, C). However, reports also suggest that Vit. C at a pharmacological dose acts as a “pro-oxidant” by generating ascorbate-free radicals and H₂O₂ in extracelluar fluids, which diffuse into the cells and cause depletion of ATP, thereby causing cell death (6). The BMSC and HSPC transplanted groups when administered with a pharmacological dose of Vit. C (200 mg/kg body wt, ip.) resulted in markedly low engraftment of cells (Supplementary Fig. S9D). BMSC and HSPC transplanted wound sections depicted a similar high DHE staining as observed in the wound-control group (Supplementary Fig. S9E, Quantitation of DHE staining-Supplementary Fig. S9F), suggesting no beneficial effect of pharmacological administration of Vit. C on reduction of ROS at the wound bed during wound healing. Thus, preconditioning of BMSCs and HSPCs with low concentrations of H₂O₂ before transplantation enhances the antioxidant defense system in the cells and makes them better equipped to survive the harsh wound environment.

OPEN IN VIEWER

FIG. 7.

Low concentration of H₂O₂-mediated preconditioning of BMSCs or HSPCs increased engraftment during transplantation. (A) Percent of male cells at the wound site when naive or preconditioned BMSCs or HSPCs were transplanted into female mice. Simultaneous treatment of Vit. C during preconditioning of BMSCs or HSPCs reversed the effect. (B) Representative images of fluorescence microscopy at lower magnification (4 × ) of regenerated wound tissue sections depicting increased co-staining of catalase GFP in preconditioned BMSCs and HSPCs as compared with naive BMSCs and HSPCs, respectively, transplanted (Tx) or wound-control groups. (C) Representative confocal photomicrographs of GFP-DHE co-stained regenerated wound tissue sections of various treated groups depicting an increased engraftment of preconditioned GFP-expressing BMSCs or HSPCs led to a marked decrease in DHE staining, indicating decreased ROS levels at the wound site. Similarly, Vit. C treatment to preconditioned BMSCs or HSPCs mitigated this effect. (D) Quantitation of confocal images using NIH Image J software depicted decreased DHE staining in preconditioned BMSCs or HSPCs transplanted groups that were inhibited in the presence of simultaneous pretreatment of Vit. C (Wd-Cont - Wound Control, PC - Preconditioned; n = 3 replicates per wound, N = 5 mice/group; *p < 0.05 as compared with wound control; **p < 0.01 as compared with wound control). DHE, dihydroxy ethidium. To see this illustration in color, the reader is referred to the web version of this article at www.liebertpub.com/ars

Preconditioned BMSC and HSPC enhance proliferation during wound tissue regeneration

Next, to evaluate proliferation of the transplanted GFP⁺ cells, we performed co-staining with Ki67 (proliferation marker) for both BMSCs and HSPCs. At postsurgery day 7, an increased co-staining of GFP/Ki67 in preconditioned BMSCs and HSPCs transplanted groups was observed as compared with naive stem cells transplanted, respectively, and wound-control groups (Fig. 8A). Separately, we have also preconditioned both BMSCs and HSPCs with high oxidative stress (100 μM H₂O₂) but it led to a decrease in proliferation (data not shown) at the wound site. These observations suggest the therapeutic relevance of 10 μM H₂O₂ but not high concentration-induced oxidative stress in a translational outcome of BMSC and/or HSPC-mediated wound tissue repair/regeneration.

OPEN IN VIEWER

FIG. 8.

Increased proliferation of preconditioned BMSCs and HSPCs improves wound healing. Representative fluorescence microscopy images at lower magnification (4 × ) of wound control, naive, or 10 μM H₂O₂ preconditioned BMSCs and HSPCs. Regenerated wound tissue section at postsurgery day 7 was co-stained with GFP and Ki67 (Insets). Preconditioned BMSCs or HSPCs transplanted wound tissues depicted a significantly high number of positively co-stained cells, suggesting higher proliferation of transplanted (Tx) cells as quantitated by intensity correlation analysis-NIH image J software and represented graphically (Wd-Cont - Wound Control, PC - Preconditioned; n = 3 replicates per wound, N = 5 mice/group; p < 0.05 as compared with *wound control; ^#naive BMSCs/HSPCs). To see this illustration in color, the reader is referred to the web version of this article at www.liebertpub.com/ars

Preconditioned BMSC but not HSPC enhance re-epithelialization during wound tissue regeneration

Further, to determine whether he transplanted stem cells at the wound site could contribute to keratinocyte generation, a co-immunostaining with cytokeratin (CK) and CD29 (BMSC) or CD11b (HSPC) was performed. At postsurgery day 7, a significantly high number of CD29 and CK co-stained cells were observed in preconditioned BMSCs as compared with naive BMSCs transplanted groups or wound-control groups (Fig. 9A left panel, Quantitation of co-localization-Fig. 9B). However, naive or preconditioned HSPCs did not depict similar CD11b and CK co-stained cells (Fig. 9A right panel, Quantitation of co-localization-Fig. 9C). To further confirm whether transplanted cells contributed to keratinocyte generation, we transplanted GFP-labeled cells and performed a co-staining of GFP and CK. At postsurgery day 7, preconditioned BMSCs transplanted wounds depicted a significant increase in double-positive (both GFP-green and CK-red) staining as compared with BMSCs transplanted groups or wound controls (Fig. 9D, quantification-9E, lower magnification images-Supplementary Fig. S10A, quantification of co-localization-Supplementary Fig. S10B). However, preconditioned HSPCs transplanted wounds did not depict co-localization of GFP-CK-expressing cells (Fig. 9D, E). These observations suggest that BMSCs but not HSPCs contribute to an increased number of keratinocytes, thereby leading to cutaneous regeneration.

OPEN IN VIEWER

FIG. 9.

Improved keratinocyte generation by preconditioned BMSCs but not HSPCs accelerates wound healing. (A) Representative confocal images depicting an increased co-staining of preconditioned BMSCs for CD29-CK (cytokeratin, a keratinocyte marker) as compared with naive BMSC transplanted (Tx) groups but not by the naive and preconditioned HSPCs transplanted groups. Quantification of the positively stained cells depicting a significant increase in BMSC transplantation group (B) but not in HSPC transplantation group (C) as compared with the respective wound-control group. (D) GFP-expressing preconditioned BMSCs or HSPCs confirmed the observation of enhanced keratinocyte generation by PC-BMSCs as compared with naive BMSCs or the wound control, whereas HSPCs transplanted (Tx) groups did not co-stain for GFP-CK. (E) Quantification of confocal images depicting an increase in the number of positively co-stained cells in various groups using intensity correlation analysis-NIH Image J software (Wd-Cont - Wound Control, PC - Preconditioned; n = 3 replicates per wound, N = 5 mice/group; *p < 0.05 and **p < 0.01 as compared with wound control). To see this illustration in color, the reader is referred to the web version of this article at www.liebertpub.com/ars

Preconditioned BMSC and HSPC enhance vascularization during wound tissue regeneration

Finally, neo-vascularization, a crucial step for tissue regeneration, was evaluated by using immunofluorescence staining of CD31 and α-SMA at the regenerated tissue sections. Both preconditioned BMSCs and HSPCs transplanted wounds depicted an increased number of endothelial cells at the wound site as compared with naive BMSCs or HSPCs as well as the wound control at postsurgery day 7 (Fig. 10A, Quantitation of co-localization-Fig. 10B). Further, to evaluate the participation of these transplanted cells in vasculogenesis, tissue sections were co-stained with GFP and CD31. Preconditioned HSPCs transplanted groups depicted an increased co-staining of GFP-CD31 as compared with naive HSPCs or the wound control, indicating the involvement of HSPCs in neo-vascularization (Fig. 10C, Quantitation of co-localization-Fig. 10D, lower magnification images-Supplementary Fig. S10C, quantification of co-localization-Supplementary Fig. S10D). Preconditioned BMSCs transplanted wound sections depicted an increased CD31 staining as compared with naive BMSCs or the wound control but the GFP-expressing cells were not fully co-localized with CD31-expressing cells, suggesting the paracrine effect of BMSCs that leads to enhanced vascularization at the wound bed. This indicates the plausible role of these transplanted preconditioned stem cells in potentiating neo-vascularization to enhance wound tissue regeneration.

OPEN IN VIEWER

FIG. 10.

Enhanced neo-vascularization by preconditioned BMSCs and HSPCs improves wound healing. (A) Representative confocal images depicting an increased co-staining of CD31-α-SMA in preconditioned BMSCs or HSPCs transplanted (Tx) groups as compared with naive BMSCs or HSPCs transplanted groups, respectively, suggesting neo-vascularization during wound tissue regeneration. (B) Quantification of confocal images using intensity correlation analysis-NIH Image J software depicting an increase in the number of positively co-stained cells in various groups. (C) Representative confocal images of GFP-expressing PC-HSPCs depicted an increased co-staining of GFP-CD31 in preconditioned HSPCs as compared with naive HSPCs, indicating neo-vascularization at the regenerated wound tissue. GFP-expressing PC-BMSCs too depicted increased staining of CD31 as compared with naive BMSCs, but all GFP-expressing cells did not co-localize with CD31, indicating a plausible paracrine effect of BMSCs in increasing vascularity in regenerated wound tissue (Wd-Cont - Wound Control, PC - Preconditioned; n = 3 replicates per wound, N = 5 mice/group; *p < 0.05 and **p < 0.01 as compared with wound-control). To see this illustration in color, the reader is referred to the web version of this article at www.liebertpub.com/ars

Exposure to H₂O₂

BMSCs and HSPCs seeded at a different seeding density depending on the assay type were treated with vehicle or 1, 10, and 100 μM of H₂O₂/well. Separately, BMSCs and HSPCs were incubated with equimolar concentrations of (10 μM) H₂O₂ along with antioxidants, Vit. C or DHA (45). In a different set of experiments, BMSCs and HSPCs were also incubated with 10 μM H₂O₂ along with signaling pathway blockers. Further, to evaluate the loss or gain function of catalase on the cells in the presence of H₂O₂, both molecular and pharmacological interventions were applied. Cells were treated with either 3 Amino Triazole (3-AT, a catalase inhibitor) or transfected with siRNA specific toward mouse catalase gene for evaluating the effect of catalase inhibition. Similarly, cells were additionally treated with PEG-catalase to evaluate the effect of exogenous catalase (32).

Cell proliferation by colony formation, growth curve, and BrdU assay

For assessing cell expansion, BMSCs and HSPCs were plated for colony formation assay, bromodeoxyuridine (BrdU) cell proliferation assay, and growth curves at different seeding densities as described earlier [10]. For colony formation assay, BMSCs and HSPCs were cultured in complete Mesencult and Methocult medium (StemCell Technologies), respectively (16, 26). DNA synthesis was measured by using a BrdU Cell Proliferation Kit (Roche) by following the manufacturer's protocol (20). For growth curve assay, BMSCs and HSPCs were seeded at a cell density of 3 × 10³ cells per well in at least triplicate with treatments as described earlier (8, 34).

Quantitative real-time PCR

Overall, 1.0 μg of total RNA extracted from treated BMSCs and HSPCs was used for cDNA synthesis from different treatment groups by using Oligo (dT) primer of First-Strand cDNA Synthesis kit (Thermo Scientific) and mouse-specific forward and reverse primers of antioxidant and cell cycle genes (Supplementary Table S1) for quantitative real-time RT-PCR (qRT-PCR) by using Applied Biosystems Step-One Plus PCR System (Applied Biosystems) as described earlier (7). Amplification of mouse-specific eukaryotic 18S rRNA and β-Actin was used in the same reaction of all samples as an internal control. All the target gene expression was normalized to eukaryotic 18S rRNA mRNA and shown as the fold change (7).

Intracellular and extracellular ROS detection

Intracellular production of ROS was detected by fluorescence probe 2′,7′-dichloro-fluorescin diacetate (DCFH-DA; Sigma-Aldrich) as previously described (49). Briefly, after treatment for 24 h or 48 h, cells were washed 3 × with PBS and then incubated in 10 μM DCFH-DA for 30 min at 37°C. The fluorescence intensity of DCF (2′, 7′-dichlorofluorescein) was determined on a Perkin Elmer Spectrofluorometer at excitation and emission wavelengths of 480 and 538 nm, respectively, and background fluorescence was subtracted by using appropriate controls. For extracellular ROS detection, Hydrogen Peroxide Cell-based assay kit was used according to the manufacturer's protocol (Cayman Chemicals). After 24 h and 48 h of treatment, fluorescence intensity was measured at 530 nm (excitation wavelength) and 590 nm (emission wavelength).

Total glutathione measurements

To measure the intracellular oxidative status of stem cells, the reduced glutathione (GSH) content of each cellular protein sample was measured according to the manufacturer's protocol (Cayman Chemicals).

Biochemical assay for catalase, SOD, and GPX

Antioxidant responses of BMSCs and HSPCs were evaluated by measuring the activity of antioxidant enzymes: catalase, SOD, and GPX. SOD and GPX activity was measured according to the protocols provided in the Cayman assay kit (Cayman Chemicals). Catalase activity was measured by using a modified protocol of Aebi.

Immunoblot analysis

Cellular protein was extracted from BMSCs and HSPCs treated with H₂O₂ alone or in different treatment regimes as described earlier, was subjected to separation by using sodium dodecyl sulfate-poly acrylamide gel electrophoresis, transferred to PVDF membranes (Amersham Biosciences), and immunoblotted by using antibodies against phosphorylated and pan signaling markers: ERK-1/2, AKT, JNK, and p38 MAPK (ThermoScientific). Signals were detected by using the chemiluminescence ECL detection system (Merck Millipore), and visualization was performed by using G: Box (Syngene, Cambridge, United Kingdom). Quantification of protein bands was performed by using NIH Image J software (7).

Cellular immunofluorescence staining—FOXO3a

Immunofluorescent staining was performed according to a reported procedure (13). In brief, treated cells were washed with cold PBS, fixed with 4% formaldehyde for 10 min at 37°C, permeabilized with 0.5% Triton X-100 for 5 min at room temperature, and fixed with methanol for overnight at 4°C. After washing with PBS, the cells were blocked with 1% BSA for 1 h and incubated with primary antibody FOXO3a in 1% BSA for 2 h at room temperature. Further, secondary anti-rabbit PE conjugated antibody was added. Nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI) contained in the mounting medium (Amresco). Confocal images were obtained with an Olympus FV1000 confocal microscope (Olympus Corp.) by using the 40 × oil immersion lens. NIH Image J software was used to quantitate nuclear localization.

SiRNA silencing—catalase

For catalase knockdown, cells were transfected with siCAT or scrambled negative control according to the manufacturer's instructions by using lipofectamine 2000 (Invitrogen) (7). The cells were further treated for 48 h in the presence or absence of H₂O₂ and were utilized for different assays.

AKT kinase assay

BMSC and HSPCs were cultured and treated as described earlier for 48 h at 37°C. Kinase activity was measured by using the In-Cell AKT ELISA Kit (Pierce), according to the manufacturer's protocol.

Transient transfection of DN plasmids—AKT and JNK

DN AKT and JNK mutant plasmids (Addgene) were transfected transiently by using lipofectamine 3000 (Invitrogen). Cells were cultured in low serum-containing medium for 48 h followed by various treatment regimes. After 48 h, cells were trypsinised and utilized for different assays.

ChIP assay

ChIP assay was performed according to the manufacturer's protocol (Millipore). Cells were treated with different treatment regime, washed, and cross-linked with 1% formaldehyde at 37°C for 10 min. Cells were lysed in buffer containing a protease inhibitor cocktail, and they were sonicated by using Bioruptor (Diagenode) followed by preclearing with Protein A Agarose/Salmon sperm DNA slurry. Immunoprecipitation was performed overnight at 4°C with 5 μg of anti-Foxo3a antibody, and this was followed by sequential washings with low-salt immune complex buffer, high-salt immune complex buffer, LiCl immune complex buffer, and, finally, with TE buffer. DNA was eluted from the beads and recovered by phenol/chloroform extraction and ethanol precipitation. The purified DNA was used as a template for PCR amplification.

BMSC or HSPC transplantation-mediated wound healing

Bone marrow-derived BMSCs and HSPCs isolated from C57BL/6J mice as described earlier (8) were used for autologous transplantation in the wound model as naive or post in vitro treatment using a low (10 μM) concentration of H₂O₂ (preconditioned, PC) and/or pretreatment with an equimolar concentration (10 μM) of H₂O₂ and Vit. C. In a separate set of experiments, isolated BMSCs and HSPCs were labeled with GFP lentivirus for 24 h followed by replacement with fresh medium. After 48 h, the cells were selected by “puromycin-selection” and infected cells were used as naive or preconditioned as described earlier for transplantation. An excision wound splinting model was generated in 8–10 week-old C57BL/6J mice as described earlier (43). Briefly, two symmetrical full-thickness excisional wounds were created beside the midline by using a 5-mm-diameter sterile biopsy punch. Transplantation of bone marrow-derived naive or preconditioned BMSCs and HSPCs was performed by injecting intradermally (id.; 0.7 × 10⁶ cells) and on the wound surface (0.3 × 10⁶ cells) as previously described (9). A silicone splint was placed so that the wound was centered within the splint. An immediate-bonding adhesive was used to fix the splint to the skin, and Tegaderm was placed over the wounds. Animal experimentation protocols were approved by the Institutional animal ethics committee (approval No. IICT/CB/AD/25/06/2014/13 and IICT/CB/AD/26/08/13/08).

Quantification of wound closure

Representative digital images of wound healing depicting wound area at different time points postwound generation (days 0, 3, and 7) in wound-control, BMSC, PC-BMSC, HSPC, and PC-HSPC transplanted groups were taken. Time to wound closure was defined as the time at which the wound bed was completely re-epithelialized and filled with new tissue. Wound area was measured by tracing the wound margin and calculated by using an image analysis program (NIH Image J software). The results were represented as percent wound closure (1), calculated in comparison with the original size of the wound as mentioned later and described earlier (3): \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*}{ \rm { Percent } } \ { \rm { Wound } } \ { \rm { Closure = } } { \frac { 1 - ( { \rm { Wound } } \ { \rm { Area } } ) } { ( { \rm { Original } } \ { \rm { Wound } } \ { \rm { Area } } ) } } \times 100 \end{align*} \end{document}

The inside edge of the splint exactly matched the edge of the wound, so that the splinted hole was used to represent the original wound size. Mice were euthanized on postsurgery day 7; skin samples including the wound and surrounding skin were harvested by using a using a 10-mm biopsy punch.

Histological and immunofluoroscence analysis of the regenerated wound

Regenerated wound tissue samples from various treatment groups were fixed in 4% paraformaldehyde. Cross-sections of wound tissue mounted on slides were stained with hematoxylin-eosin, Masson's Trichrome, and DHE. The regenerated wound tissue was separately homogenized in respective assay buffers to evaluate catalase, SOD, and GPX assay. RNA extracted from wound tissue samples was utilized for gene expression of oxidative stress-associated genes by using qRT-PCR analysis. Separately, frozen wound tissue sections from the various groups were stained with specific antibodies against catalase, GFP, Ki67, CD29, CD11b, CK, CD31, and αSMA (Cell Signaling Technologies). Isotype control antibodies were used for negative controls. Sections were examined with Olympus FV1000 and FV10i confocal microscope. Co-localization of different cellular markers was evaluated by using the intensity correlation analysis of NIH Image J software, and data were represented as Pearson's coefficient value (1).

Vit. C in wound healing

Parenteral administration of Vit. C (250 mg/kg body wt, i.p. per day) (6) was carried out for 7 days in the excisional splinting wound model transplanted with BMSCs or HSPCs. The mice were monitored for wound healing, engraftment of transplanted cells, and histological analysis.

Statistical analysis

Data from experiments performed with repeated BMSC/HSPC isolations at least thrice were statistically analyzed by taking mean ± standard deviation (SD) or standard error of mean (SEM). To determine the difference between treatment group and their respective controls, Graph Pad prism version 6.05 was used to obtain statistical significance by using one- or two-way ANOVA followed by Tukey's multiple-comparison test. Blots and photomicrographs represent experiments reproduced at least thrice with similar results.