Cell therapy is being widely tested as a treatment for “no-option” CLTI patients in both pre-clinical and clinical stages [8, 42,43,44,45]. However, the discrepancies registered in these studies as result of differences in cell types employed, cell combination, administration routes or doses applied, make necessary to reach a consensus regarding the optimal “cellular-cocktail” for these patients, and set-up optimized but well stablished protocols prior further application into the clinic [8, 46, 47].

An important factor to address in cell therapy is the choice of cell source. In this sense, although the therapeutic potential of ECFCs and MSCs has been extensively investigated in the context of vascular regeneration [48, 49], all studies have always used ECFCs derived from umbilical cord blood (CB) or peripheral blood (PB). CB-ECFCs have proven to show great angiogenic potential, but their use still has some ethical issues due to their embryonic origin. On the other hand, PB-derived ECFCs can be obtained from adults, but they are extremely rare in adult peripheral blood [22, 50, 51]. Consequently, the success of their extraction is highly variable, making their extraction unreliable across patient populations. Alternatively, the isolation of ECFCs from AT is highly efficient [22], providing an alternative that is easily achievable, and susceptible to be applied as allogenic or autologous therapy. Finally, AT is also chosen as an optimal source of MSCs, with easier and less invasive access than the BM itself. Besides, recent studies have shown a better recovery of HLI animal models treated with AT-MSCs than with BM-MSCs [52].

In the current study we have evaluated, for the first time to our knowledge, the regenerative potential of a combination of adipose tissue (AT)-derived cells, AT-ECFCs and AT-MSCs, in a murine model of CLTI, compared to the effect of CB-ECFCs with AT-MSCs. Moreover, the molecular mechanisms altered in CLTI and in response to both combinations have been evaluated at a proteomic level.

Based on our results, FAL significantly reduced BF (> 80%), which was accompanied by inflammation, gradual presence of black nails and necrotic fingers, as well as a negative impact in the animals’ motility. Despite this, a recovery in BF could be seen by day 7, even in IC untreated mice, which might be explained by repairing mechanisms triggered in response to hypoxia an ischemic events, such as the formation of new capillaries and new muscle fibers [53], in an attempt to recover from the absence of oxygen and lack of nutrients, as previously reported [54]. Such vessel sprouting takes place during the first days post-ischemia, even in non-treated tissues [30]. Afterwards, the newly formed vessels either grow and become functional or, on the contrary, they are immature and fragile, and they might contribute to the tissue instability [55, 56]. Remarkably, by day 21, the BF recovery seen after cell administration, for instance co-administration of AT-MSCs and ECFCs (either CB or AT source), was higher than in IC untreated mice, where perfusion ratios were significantly lower, in agreement with previous studies [49, 57]. Besides, although the highest BF ratios were observed with the AT combination on days 7 and 14, compared to the CB group, by day 21 the recovery was similar with both sets of cells. Furthermore, our data indicated that the transplantation of AT-ECFCs seemed to foster BF reperfusion by promoting a higher number of functional vessels, while in CB-ECFCs treated mice, such improvement appeared associated to an increase of lumen vessel diameters, as further corroborated by proteomic results. The fact that we did not detect any traces of human cells after 21 days post-injection, suggests that ECFCs and MSCs might exert their long-term angiogenic effect in a paracrine fashion, in agreement with previous findings [5, 21, 30, 58,59,60], with many studies reporting the presence of these cells in the ischemic muscles up to 14 days [57, 61].

Overall, the LFQ proteomic analysis corroborated previous results from our lab regarding the long-term effect of the ischemic process [30], with an important number or protein changes detected in the ischemic limbs that reflected an enhanced immune and inflammatory response, as well as changes associated to vascular-related processes and muscle damage itself, which was accompanied, among others, by an alteration of energy homeostasis, oxidative stress, mitochondrial dysfunction or dysregulation of lipid metabolism, as further disclosed above.

Proteomic changes related to vascular damage in ischemic mice

Based on functional classification analyses, the differential protein profiles identified represented an up-regulation of vessel occlusion, as expected, as well as proteins associated with an activation of angiogenesis and vasculature development. Among the proteins up-regulated in all ischemic mice (IC, CB and AT) we found Angiopoietin-like Protein 2 (ANGPTL2), a well-known pro-angiogenic agent that appears to enhance ECFCs vasculogenesis in vitro [62] and whose expression has been associated with increased number of blood vessels in mice [63, 64]. Similarly, up-regulation of the Low-density lipoprotein (LDL) receptor-related protein-1 (LRP1), might be part of a repairing mechanism activated in ischemic tissues, given the role of LRP1 in the regulation of several signaling pathways responsible for maintaining and restoring vascular homeostasis [65]. Likewise, the Amyloid precursor protein (APP), associated to amyloid plaque formation and Alzheimer, has been found to promote angiogenesis, stimulating ECs migration, proliferation and formation of new vessels, at least in vitro [66]. Besides, the Platelet-derived growth factor receptor beta (PDGFRB), was also found up-regulated in ischemic mice. This pro-angiogenic factor is secreted from the endothelium of angiogenic sprouts, attracting comigration pericytes [67,68,69], and it also stimulates vascular smooth muscle cells proliferation and induces mural cell fate in MSCs [23, 69, 70].

Some of the proteins found in ischemic mice, however, were differentially expressed in cell-treated mice compared to untreated ones, correlating with a higher activation of angiogenic processes in CB or AT mice than in IC 21 days post-FAL. Among them, the Adenosine triphosphate (ATP)-binding cassette (ABC) transporter ABCC9, also known as sulfonylurea receptor 2 (SUR2), is a component of the ATP-sensitive potassium (KATP) channel responsible for ATP-dependent inward K+ transport [71]. KATP channels have been identified as important mediators of ischemic preconditioning, limiting ischemic damage [72], and they also constitute key modulators of vascular smooth muscle activity by regulating blood pressure and episodic coronary artery tone [73]. Remarkably, ABCC9 was down-regulated in IC untreated mice, while its levels were restored with CB and more significantly with AT cells. This might represent an alteration of the K + transport through KATP channels after ischemia which was somehow restored in response to cell treatment, and for instance, a better recovery of the vascular tone as well in these mice.

Immune response and altered metabolism in ischemic mice

In terms of inflammation, numerous proteins were found up-regulated in ischemic mice compared to shams. For example, the complement C5 protein has been widely associated to ischemia reperfusion injury or activation of neutrophils [74,75,76], as well as with the progression of atherosclerotic lesions [77]. Similarly, Galectin-3 (LGALS3) participates in the modulation of the inflammatory response as an important mediator of the reparative process driven by macrophages, and it has been associated with ischemia injury progression as well [78, 79]. Finally, other proteins up-regulated in ischemic mice were the S100A8/S100A9 complex, an important component in neutrophil mobilization [30, 80], or the transmembrane protein beta-2 integrin (ITGB2), which is expressed exclusively in leukocytes after an activation signal, promoting neutrophil adhesion strengthening, cell spreading and crawling [81,82,83]. The effects from the femoral blockade and hypoxic conditions were also reflected by alteration of the glucose metabolism (i.e. ALDOA, GAPDHS, PFKP, PGAM2, SDHB), with many pathways related that appeared down-regulated even at the long-term (glycolysis, gluconeogenesis and oxidative phosphorylation steps). All these processes are required to maintain energy balance and normal physiological function [84, 85]. Similarly, energy imbalance was represented by down-regulation of proteins affecting mitochondrial function integrity and mitochondrial respiration (TFAM, SOD1, SOD2, MFN1, BCS1L, UQCR11) [86,87,88], as well as proteins related to fusion/fission mitochondria processes and mitophagy (SEPTIN2, OPA1, BAX, BAK1, ATG3, AIFM1), while there was an up-regulation of proteins associated to mitochondrial degradation (ADH5, ADIPOQ, AGT, AIFM1, APP, ATG3, BAK1, BAX, DNM1L, etc.) [89,90,91,92,93]. On the other hand, glucose imbalance was accompanied by up-regulation of lipid metabolism, both lipid synthesis and lipid hydrolysis (FABP5, FASN, HEXA, LIPE, PLIN1, etc.). Cell administration, however, reverted the levels of some energy related proteins down-regulated in IC. For instance, proteins like PGC-1/ERR-induced regulator in muscle 1 (PERM1), were up-regulated in cell-treated mice compared to IC. PERM1 is a striated muscle-specific regulator of mitochondrial bioenergetics, required for the high ATP demanding contractile activity of muscle [94]. Also, in cardiomyocytes, PERM1 protects against stress-induced cellular damage because of hypoxia [95]. Other proteins like the Ubiquinol-cytochrome c reductase (UQCR11) decreased significantly in IC and AT mice, but their levels recovered in the presence of CB combination. UQCR11 is a core subunit of mitochondrial complex III, an essential component of the mitochondrial electron transport chain. Thus, although further studies should validate these results, at least in this case CB appeared to be beneficial to restore mitochondrial function and therefore to promote energy restoration in the ischemic muscles.

Modulation of necrotic processes in cell-treated mice

Other common pathways associated to ischemia such as apoptosis, necrosis and cell death were found up-regulated in ischemic mice (both, IC untreated and AT or CB treated ones) although, based in our data, cellular administration somehow modulated and reduced the necrotic process, compared to IC untreated mice. Indeed, according to the necrotic scores applied, IC mice presented the highest percentage of necrotic tissue along the assay, while the necrotic ratios in CB or AT mice, although high, were slowed down, mainly in the presence of CB-ECFCs. Moreover, the CB group had the slowest ischemic progression along the assay. Interestingly, together with necrosis and apoptosis, a significant upregulation of ferroptosis, an important regulatory mechanism that induces skeletal muscle cell death and prevents skeletal muscle proliferation and differentiation [40], was also reported in all ischemic mice. Indeed, the changes seen for several proteins related to this process (ACSL4, TFRS, GPX4, etc.) [38, 39, 41], as well as the increase seen for the final product of lipid peroxidation, MDA [96], were indicative of ferroptosis up-regulation after 21 days of FAL induced ischemia. These results indicated that cell administration, at least in the case of ferroptosis, was not sufficient to restrain this process.

Ameliorated immune response in cell treated mice

Similarly, despite the up-regulation seen of immune related pathways in response to ischemia, one of the main achievements of cell administration was, according to the proteomic analysis, a general down-regulation of inflammatory and immune response processes in AT/CB treated mice compared to IC untreated ones. These results were validated by IHC, detecting a higher number of macrophages and neutrophils in IC ischemic muscles 21 days post-FAL, compared to non-ischemic shams, but also to cell-treated mice. These results agreed with previous findings from our lab after the administration of CACs [30], and moreover, with studies reporting that a combination of ECFCs and MSCs from adipose tissue modulated and reduced neutrophils recruitment and activation in grafts implanted into mice as the vasculature matured [97]. Remarkably, the AT group presented the lower number of both, macrophages and neutrophils, in the injured tissue. Thus, a decrease in macrophage activation, neutrophil trafficking, adhesion capacity and chemotaxis was observed, with proteins down-regulated (ALOX5AP, CD14, CORO1A, CTSB, FCER1G, IL1RN, LBP) [98,99,100,101,102,103,104,105] and up-regulated (NQO1, PROCR, TIPE2) in AT and CB mice compared to IC ones, as it was also previously reported [106,107,108]. Overall, our results indicated that the combination of AT-MSC with either AT- or CB-ECFCs alleviated the impact of the negative effects of ischemia, reducing inflammation but also tissue necrosis.

Muscle regeneration and tissue repair

In terms of muscle damage and tissue repair, many proteins associated either with muscle degradation, fibrosis, or necrosis of the skeletal muscle itself were identified, but also proteins involved in muscle regeneration. In this regard, based on the proteomic results, the administration of cells reinforced the repairing processes, with down-regulation of fibrosis and muscle damage (more intense in CB treated mice than AT ones) compared to IC, and up-regulation of proteins that participate in the reorganization and remodeling of muscle tissue fibers.

Among the DEPs identified, a critical protein found was Cysteine and Glycine Rich Protein 3 (CSRP3, also known as muscle LIM protein or MLP), which participates in muscle differentiation and maintenance of the contractile apparatus. Moreover, CSRP3 might activate autophagy in response to acute starvation and hypoxia. Autophagy is required to preserve the myotube structure and function during myotube formation and thereby prevents alteration of myotube morphology or integrity [109]. Therefore, the up-regulation of CSRP3 in AT and CB mice compared to IC might represent a positive effect of these cells, first to prevent the accumulation of damaged materials inside the cells, and second to promote muscle differentiation and repair.

Vesicle-associated membrane protein 2 (VAMP2, also known as synaptobrevin-2), which was found up-regulated only in response to cell treatment (AT and CB vs SH and IC), is involved in membrane trafficking, with a potential role in muscle regeneration [110, 111]. Initially identified in synaptic vesicles from rat brain [112], VAMP2 has been also detected in quiescent muscle satellite cells. Moreover, VAMP2 has been found upregulated in immature myofibers during muscle regeneration, in correlation with the release of VEGF, IGF and other growth factors by these cells.

Another important protein required to ensure proper functioning of the muscles is creatine kinase (CKM), an enzyme that catalyzes the reversible transfer of γ-phosphate from ATP to creatinine to produce phospho-creatine (PCr) and ADP, ensuring an efficient cytosolic storage of high-energy phosphates for rapid, focal ATP replenishment. The Cr/CK system provides an indispensable support to maintain cellular homeostasis when imbalances in ATP supply and demand [113]. In response to ischemia, a significant down-regulation of CKM was seen in our mice even 21 days post-FAL, what is associated to down-regulation of PCr as well as ATP levels, and reduction of energy availability, essential for the correct functioning of the muscle. Remarkably, although CKM levels were still lower in cell treated mice than control shams, we could see in AT and more significantly CB treated mice, a significant recovery of CKM levels compared to IC, which might have also contributed to a better recovery of ischemic muscles by a higher availability of energy resources.

CB vs AT response

Finally, in terms of the cell combinations applied, either CB- or AT-ECFCs and AT-MSCs, both seemed equally efficient in promoting revascularization of the ischemic tissues, since they both achieved a significant perfusion recovery at the long-term, with modulation of the inflammatory process and reduction of the necrotic progression, compared to IC untreated mice. Nevertheless, while BF recovery seemed initially faster with the AT combination, probably because these cells induced angiogenesis and formation of a higher number of vessels, the recovery with CB was similar after 21 days, enhancing reperfusion by the development of wider vessels. Besides, the CB group had the slowest ischemic progression along the assay. In terms of the molecular mechanisms altered in response to ischemia, slight differences were seen between AT and CB treated mice. Several proteins had different tendencies in these groups and even presented statistically significant up- or down-regulation between them, although the overall response was similar. Nevertheless, in the AT group there was up-regulation of ROS production, as well as slightly higher muscle damage compared to CB, where fibrosis was also less pronounced, according to our functional analysis, than in AT treated mice. For example, the mitochondrial protein NAD-dependent deacetylase sirtuin-3 (SIRT3) was down-regulated in IC and AT mice compared to Shams, but its levels recovered in CB mice. Remarkably, SIRT3 is known to protect against ROS-induced damage by reducing ROS and influences many energy metabolism processes through deacetylation of key enzymes [114].