The secretome of stressed peripheral blood mononuclear cells increases tissue survival in a rodent epigastric flap model

Abstract Reconstructive surgery transfers viable tissue to cover defects and to restore aesthetic and functional properties. Failure rates after free flap surgery range from 3 to 7%. Co‐morbidities such as diabetes mellitus or peripheral vascular disease increase the risk of flap failure up to 4.5‐fold. Experimental therapeutic concepts commonly use a monocausal approach by applying single growth factors. The secretome of γ‐irradiated, stressed peripheral blood mononuclear cells (PBMCsec) resembles the physiological environment necessary for tissue regeneration. Its application led to improved wound healing rates and a two‐fold increase in blood vessel counts in previous animal models. We hypothesized that PBMCsec has beneficial effects on the survival of compromised flap tissue by reducing the necrosis rate and increasing angiogenesis. Surgery was performed on 39 male Sprague–Dawley rats (control, N = 13; fibrin sealant, N = 14; PBMCsec, N = 12). PBMCsec was produced according to good manufacturing practices (GMP) guidelines and 2 ml were administered intraoperatively at a concentration of 2.5 × 107 cells/ml using fibrin sealant as carrier substance. Flap perfusion and necrosis (as percentage of the total flap area) were analyzed using Laser Doppler Imaging and digital image planimetry on postoperative days 3 and 7. Immunohistochemical stainings for von Willebrand factor (vWF) and Vascular Endothelial Growth Factor‐receptor‐3 (Flt‐4) were performed on postoperative day 7 to evaluate formation of blood vessels and lymphatic vessels. Seroma formation was quantified using a syringe and flap adhesion and tissue edema were evaluated clinically through a cranial incision by a blinded observer according to previously described criteria on postoperative day 7. We found a significantly reduced tissue necrosis rate (control: 27.8% ± 8.6; fibrin: 22.0% ± 6.2; 20.9% reduction, p = .053 vs. control; PBMCsec: 19.1% ± 7.2; 31.1% reduction, p = .012 vs. control; 12.9% reduction, 0.293 vs. fibrin) together with increased vWF+ vessel counts (control: 70.3 ± 16.3 vessels/4 fields at 200× magnification; fibrin: 67.8 ± 12.1; 3.6% reduction, p = .651, vs. control; PBMCsec: 85.9 ± 20.4; 22.2% increase, p = .045 vs. control; 26.7% increase, p = .010 vs. fibrin) on postoperative day 7 after treatment with PBMCsec. Seroma formation was decreased after treatment with fibrin sealant with or without the addition of PBMCsec. (control: 11.9 ± 9.7 ml; fibrin: 1.7 ± 5.3, 86.0% reduction, 0.004 vs. control; PBMCsec: 0.6 ± 2.0; 94.8% reduction, p = .001 vs. control; 62.8% reduction, p = .523 vs. fibrin). We describe the beneficial effects of a secretome derived from γ‐irradiated PBMCs on tissue survival, angiogenesis, and clinical parameters after flap surgery in a rodent epigastric flap model.

(Flt-4) were performed on postoperative day 7 to evaluate formation of blood vessels and lymphatic vessels. Seroma formation was quantified using a syringe and flap adhesion and tissue edema were evaluated clinically through a cranial incision by a blinded observer according to previously described criteria on postoperative day 7. We found a significantly reduced tissue necrosis rate (control: 27 increase, p = .045 vs. control; 26.7% increase, p = .010 vs. fibrin) on postoperative day 7 after treatment with PBMCsec. Seroma formation was decreased after treatment with fibrin sealant with or without the addition of PBMCsec. (control: 11.9 ± 9.7 ml; fibrin: 1.7 ± 5.3, 86.0% reduction, 0.004 vs. control; PBMCsec: 0.6 ± 2.0; 94.8% reduction, p = .001 vs. control; 62.8% reduction, p = .523 vs. fibrin). We describe the beneficial effects of a secretome derived from γ-irradiated PBMCs on tissue survival, angiogenesis, and clinical parameters after flap surgery in a rodent epigastric flap model.

K E Y W O R D S
secretome, angiogenesis, flap surgery, necrosis, reconstructive surgery, tissue regeneration

| INTRODUCTION
Reconstructive surgery uses local, pedicled, or free flaps to restore the function and appearance after tumor resection or trauma. [1][2][3] The rate of free flap failure ranges from 3 to 7%. 4,5 Patients requiring reconstructive surgery often suffer from diabetes mellitus or peripheral vascular disease, that were shown to increase the rate of flap failure up to 4.5-fold. [6][7][8][9] This may lead to tissue necrosis and jeopardizes the reconstructive result. Necrosis can subsequently cause a series of events resulting in inflammation and further loss of tissue integrity. 10 Experimental therapies including the use of antioxidants, vasodilators, anti-inflammatory drugs, and hyperbaric oxygen led to a decrease in necrosis rates. [11][12][13][14][15][16] Alternatively, therapeutic angiogenesis through the application of pro-angiogenic factors, such as Platelet-derived growth factor (PDGF), and Vascular Endothelial Growth Factor (VEGF) has been demonstrated to improve the survival of compromised flaps by improving tissue perfusion. 10,[17][18][19][20][21] The use of platelet-rich plasma (PRP) resulted in an increase in flap survival of 20% by inducing angiogenesis and reducing the inflammatory response. 22,23 Currently, none of these therapies is used routinely during flap surgery. Fibrin sealants have been investigated for their hemostatic and adhesive properties and their ability to locally deliver and sustainably release growth factors, thus providing an important role as a biomatrix. [24][25][26][27][28] Previous studies showed that γ-irradiated, stressed peripheral blood mononuclear cells (PBMCs) represent an easily accessible source for the production of a cellular secretome (PBMCsec) with cytoprotective, regenerative, and immunomodulatory capacity, especially in ischemic tissues. 29,30 In contrast to experimental therapies applying only single growth factors, PBMCsec consists of a mixture of lipids, proteins, and extracellular vesicles that together represent the regenerative potential of this cell-free therapy. 31 Several mechanisms have already been characterized, indicating that the entire PBMCsec is required to exert its full action spectrum, as subfractions thereof did not reach the same beneficial effects. 32,33 The pleiotropic effects of the secretome components better resemble the physiologic environment of wound healing occurring in the body, thus resulting in a faster and better regeneration. PBMCsec was successfully applied in animal models of wound healing, tissue ischemia or inflammation. [34][35][36][37][38][39] The application of PBMCsec enhanced wound healing and angiogenesis in a murine full-thickness skin wound model.
In vitro investigations showed increased migration and proliferation rates of keratinocytes, fibroblasts, and endothelial cells. 35 In a porcine model of burn injury, PBMCsec significantly improved the epidermal thickness, and led to a two-fold increase in angiogenesis. 36 The safety and tolerability of topically administered autologous PBMCsec in human dermal wounds has already been proven in a clinical phase I trial (ClinicalTrials.gov Identifier: NCT02284360). 40
F I G U R E 2 (a) Flap perfusion was measured using Laser Doppler Imaging (LDI). Examples for the color-coded image after perfusion measurements are given for each group. After ligation of the unilateral neurovascular bundle, the perfusion decreases on the contralateral side. The situation appeared completely different in the ischemic zone. day 3) (Figure 2a,c, Table 1).

| Blood vessel density is increased in the PBMCsec treated flaps
We found a statistically significant increase in blood vessels in the 3.6% reduction, p = .651 vs. control) group. (Figure 3a-d,  Figure 3e-h, Table 1).

| DISCUSSION
We were able to demonstrate a reduced postoperative flap necrosis and a significantly improved rate of angiogenesis after flap surgery through the application of secretome derived from γ-irradiated, and miRNA composition of the secretome toward a regenerative phenotype. 33 In most of the studied biological processes the integrity of the entire PBMC secretome was necessary to achieve full regenerative effects as sub-fractions or single cell types did not reach the same capacity. 32,33 Unlike previously described monocausal therapies, the effects of PBMCsec can therefore be attributed to its heterogeneity, resembling the physiological environment necessary for tissue regeneration. Previously described therapies with single growth factors significantly reduced postoperative flap necrosis at a similar rate to the described results. However, these experimental therapies mainly focused on angiogenesis and therefore lack additional immunomodulatory and cytoprotective effects that may play a role in patients with co-morbidities. 26,27 In correlation with the previously described effects in animal models of wound healing we found an increased density of vWF+ blood vessels after PBMCsec treatment. 35,36 The TNF/TNFRSF1B signaling pathway was described as the mechanism underlying the γ-irradiation-induced pro-angiogenic activity of PBMCsec. 32 Clinically, the impairment of flap perfusion leads to inflammation and secondary tissue damage. 10,49 The immunomodulatory and cytoprotective effects of PBMCsec were previously described in animal models of myocarditis, contact hypersensitivity, and cerebral ischemia. 37

| Production of secretome
PBMCsec was produced according to previously described methods (LOT: 399014). 39 In anticipation of possible future clinical applications, the production followed GMP guidelines. This production process was shown to yield reproducible results. 47

| Rodent epigastric flap model
The rodent epigastric flap model was performed as described. 42 39 adult male Sprague-Dawley rats (weighing 422 ± 30 g) were

| Digital image planimetry
The surface area of the flap including the necrotic flap area was traced onto a transparent acrylic foil, which was then photographed. Digital images were analyzed using the ImageJ software. 53 (Table 2). 42 Seroma fluid between the muscular abdominal wall and the flap was aspirated with a syringe and measured. F I G U R E 6 (a) A previously described epigastric flap model was used to evaluate the regenerative and angiogenic effects of the treatment protocol. 42

| Statistical analysis
We used IBM SPSS Statistics 24.0 (IBM, Armonk, NY) for data analysis. In case of a normal distribution of a metric variable, the Student's t-test was used to compare groups. Otherwise, the nonparametric Mann-Whitney U-Test was used. If not stated otherwise, results are given as mean ± standard deviation (SD). In all calculations, a p-value <.05 was considered statistically significant. The p-values were not adjusted for multiple comparisons.

| CONCLUSIONS
In conclusion, we demonstrated a significantly reduced necrosis rate in combination with increased blood vessel density after a single, intraoperative application of the secretome derived from γ-irradiated,