Lung Transplant Immunomodulation with Genetically Engineered Mesenchymal Stromal Cells—Therapeutic Window for Interleukin-10

Lung transplantation results are compromised by ischemia–reperfusion injury and alloimmune responses. Ex vivo lung perfusion (EVLP) is used to assess marginal donor lungs before transplantation but is also an excellent platform to apply novel therapeutics. We investigated donor lung immunomodulation using genetically engineered mesenchymal stromal cells with augmented production of human anti-inflammatory hIL-10 (MSCsIL-10). Pig lungs were placed on EVLP for 6 h and randomized to control (n = 7), intravascular delivery of 20 × 106 (n = 5, low dose) or 40 × 106 human MSCs IL-10 (n = 6, high dose). Subsequently, single-lung transplantation was performed, and recipient pigs were monitored for 3 days. hIL-10 secretion was measured during EVLP and after transplantation, and immunological effects were assessed by cytokine profile, T and myeloid cell characterization and mixed lymphocyte reaction. MSCIL-10 therapy rapidly increased hIL-10 during EVLP and resulted in transient hIL-10 elevation after lung transplantation. MSCIL-10 delivery did not affect lung function but was associated with dose-related immunomodulatory effects, with the low dose resulting in a beneficial decrease in apoptosis and lower macrophage activation, but the high MSCIL-10 dose resulting in inflammation and cytotoxic CD8+ T cell activation. MSCIL-10 therapy during EVLP results in a rapid and transient perioperative hIL-10 increase and has a therapeutic window for its immunomodulatory effects.


Introduction
Lung transplantation can return a patient with severe end-stage lung disease to normal life.However, the median survival of lung transplant recipients is currently limited to about 7 years [1][2][3], mainly due to primary graft dysfunction, acute rejection and chronic lung allograft dysfunction [4][5][6].These factors involve an interplay between innate and adaptive immunity [4,7,8].Novel donor lung immunomodulatory therapeutic strategies could limit early pathological immune responses and improve long-term survival.
Ex vivo lung perfusion (EVLP) is an innovative approach to assessing donor lung quality before lung transplantation [9].EVLP has safely increased the number of lung transplants, as marginal donor lungs with stable lung function during EVLP can be transplanted with excellent early [10], mid-term and long-term results [11].EVLP is also an ideal platform for novel therapeutics as, for example, gene and cell therapies could be administered to the donor lung during the ex vivo perfusion period [12][13][14].
Gene therapy has been successfully used in experimental transplant and EVLP models [13,15,16].However, as it takes several hours to achieve therapeutic levels of the transgenic protein, donor lung gene therapy could miss the relatively narrow therapeutic time frame for perioperative graft protection [15,16].Various cell therapies such as mesenchymal stromal cells (MSCs) have also been applied during EVLP [13,17].MSCs elicit immunomodulatory effects primarily through paracrine mechanisms [17][18][19], and tailored genetic engineering of MSCs further expands their therapeutic potential [14,18].We recently generated genetically modified MSCs that have augmented anti-inflammatory interleukin-10 production (MSCs IL-10 ) [20].Treatment of human lungs rejected from clinical transplantation with MSCs IL-10 during EVLP efficiently elevated IL-10 levels within minutes [20].
Here, we used a translational pig EVLP and lung transplant model to determine the potential protective and immunomodulatory effects of MSC IL-10 therapy, with two different MSC IL-10 doses to investigate dose-related effects.Administration of MSCs IL-10 during EVLP increased IL-10 levels rapidly and dose-dependently and resulted in transient IL-10 elevation after lung transplantation.Interestingly, the two MSC IL-10 doses induced divergent immunological effects, with the low MSC IL-10 dose resulting in decreased apoptosis and lower lung macrophage activation, while the high MSC IL-10 dose caused increased inflammation and cytotoxic CD8 + T cell activation.

Materials and Methods
Human umbilical cord MSCs with augmented human IL-10 production (hIL-10) were genetically engineered by adenoviral transduction and then cryopreserved.hIL-10 secretion was confirmed in vitro.Pig double lungs subjected to 24 h cold ischemia were connected to clinical-grade EVLP (using the Toronto EVLP technique) [21] for 6 h and randomized to control (n = 7), or to receive 20 × 10 6 (n = 5) or 40 × 10 6 cryopreserved MSCs IL-10 (n = 6) through the pulmonary artery in a blinded manner.Lung function parameters, perfusate, lung tissue and bronchoalveolar lavage (BAL) samples were collected during EVLP.After EVLP, the left lung was transplanted into a recipient pig that received methylprednisolone and cyclosporine A background immunosuppression [22].Three days after transplantation, graft evaluation was performed.Blood, tissue and BAL samples were collected for biochemical, histological and immunological analysis.
Detailed n-values are given in Supplementary Table S1 and in the respective Figure Legends.Please see the online supplement for detailed methods.All animals received humane care in compliance with the "Principles of Laboratory Animal Care," formulated by the National Society for Medical Research, and The Guide for the Care of Laboratory Animals, published by the National Institutes of Health.The experimental protocol was approved by the Animal Care Committee of the Toronto General Hospital Research Institute (AUP 2999).
To investigate the potential immunomodulatory effects of MSC IL-10 therapy during lung transplantation, a pig EVLP and transplant model was used (Figure 1C).MSCs IL-10  were administered during EVLP through the pulmonary artery (Figure 1D) 71 ± 4 min after MSC IL-10 thawing.Measurements of the supernatant of the MSC IL-10 preparation indicated the generation of 0.57 ± 0.14 and 1.66 ± 0.41 µg of preformed hIL-10 (mean ± SD) by the 20 × 10 6 and 40 × 10 6 MSCs IL-10 , respectively, that was thus co-injected into the EVLP circuit together with the MSCs IL-10 (Figure 1E).IL-10 (n = 6).Subsequently, the left lung was transplanted into a recipient pig that received methylprednisolone and cyclosporine A background immunosuppression, and was sacrificed 3 days after transplantation.(D) A clinical-grade Toronto ex vivo lung perfusion system was used for the EVLP experiments, and MSCs IL-10 were administered to the EVLP circuit in a blinded and randomized manner with a syringe (white arrow) and a tubing extension was attached to the  IL-10 (n = 6).Subsequently, the left lung was transplanted into a recipient pig that received methylprednisolone and cyclosporine A background immunosuppression, and was sacrificed 3 days after transplantation.(D) A clinical-grade Toronto ex vivo lung perfusion system was used for the EVLP experiments, and MSCs IL-10 were administered to the EVLP circuit in a blinded and randomized manner with a syringe (white arrow) and a tubing extension was attached to the pulmonary artery cannula (black arrow).(E) Analysis of preformed hIL-10 in MSC IL-10 supernatant taken immediately prior to MSC IL-10 administration.Data mean ± standard deviation and analyzed with (B) two-way ANOVA using Šidák correction for multiple comparisons or (E) 2-tailed Student's t-test.Ad, adenovirus; EVLP, ex vivo lung prefusion; hIL-10, human interleukin-10; MSC, mesenchymal stromal cell; TX, transplantation.

MSC Delivery during EVLP Results in Rapid hIL-10 Elevation in Perfusate, Lung Tissue and Airways
To investigate the kinetics of hIL-10 production during EVLP, pig double lungs with 24 h cold ischemia were placed on the EVLP system, and 20 × 10 6 or 40 × 10 6 MSCs IL-10 were administered to the perfusate entering the pulmonary artery (Figure 2A) after the first hour on EVLP (Supplementary Table S1).Perfusate hIL-10 levels increased within 5 min after cell administration to 492 ± 144 and 1022 ± 390 pg/mL in the 20 × 10 6 and 40 × 10 6 MSC IL-10 groups, respectively (Figure 2B), likely resulting from the co-administrated preformed hIL-10 (Figure 1E) being distributed into the EVLP circuit.Subsequently, perfusate hIL-10 concentration continued to increase steadily and dose-dependently and reached very high levels of 6068 ± 1384 and 17,194 ± 5573 pg/mL in the 20 × 10 6 and 40 × 10 6 MSC IL-10 groups, respectively, 6 h after EVLP start (Figure 2B).
A corresponding increase in lung hIL-10 was observed with tissue concentrations reaching 37 ± 16 and 76 ± 44 pg per mg of lung protein at the end of EVLP in the 20 × 10 6 and 40 × 10 6 MSC IL-10 groups, respectively (Figure 2C), far exceeding the levels achieved in previous adenoviral gene therapy studies [15].Similar hIL-10 levels were detected in different anatomical lung areas (Figure 2D), and an analysis of BAL samples indicated an increase in hIL-10 in the airway compartment (Figure 2E).IL-10 treatment during EVLP results in rapid dose-dependent hIL-10 elevation that decreases after lung transplantation.Pig double lungs subjected to 24 h cold ischemia were (A) connected to EVLP for 6 h, 20 × 10 6 (n = 5) or 40 × 10 6 (n = 6) MSCs IL-10 were administered, and hIL-10 was measured using ELISA.(B) After MSC IL-10 administration, EVLP perfusate hIL-10 levels were rapidly and dose-dependently elevated within minutes and continued to steadily increase for the duration of EVLP.(C) Lung tissue hIL-10 protein levels increased during EVLP, and (D) no significant differences were detected between samples taken from the upper or lower lobes, or from superficial or deep areas of the lung.(E) BAL hIL-10 was dose-dependently elevated 6 h after EVLP start.(F) After EVLP, the left lung was transplanted to a recipient pig that received methylprednisolone and cyclosporine A immunosuppression. (G) Recipient plasma hIL-10 protein levels peaked dose-dependently during the first 4 h after transplant reperfusion and declined to low levels within 1-2 days after transplantation.(H) Lung tissue hIL-10 protein levels were elevated 1 h after lung transplant reperfusion but were undetectable at 3 days.(I) Quantitative RT-PCR with human-specific primers of lung samples showed markedly increased hIL-10 mRNA expression at the end of EVLP, lower hIL-10 mRNA levels 1 h after transplant reperfusion and undetectable levels 3 days after transplantation.(J) Quantitative RT-PCR for human VEGF mRNA revealed low, but detectable levels of human VEGF mRNA in the lung 6 h after EVLP and 1 h after transplantation.(K) Lung tissue multiplex immunofluorescent and fluorescent in situ hybridization staining with human-(HLA Class I and ALU) and transgene-specific markers (FLAG tag; attached to the hIL-10 transgene) detected human MSCs IL-10 in the pig lung 6 h after EVLP start.(K) MSCs IL-10 co-expressed ALU, HLA Class I and FLAG, were relatively large, and (K,L) resided in duplicate or as (M,N) single cells.(L) Co-staining with endothelial (CD31) and epithelial (pan cytokeratin) markers indicated that MSCs IL-10 mainly localized in the (K) lung interstitium and occasionally in the (L) alveolar space.(O) Quantitative PCR with ALU primers indicated the presence of human cells in the pig lungs at the end of EVLP, and 1 h after lung transplantation, but did not detect human genomic DNA in the transplant, or in any non-transplant recipient tissues 3 days after lung transplantation.Data mean ± standard deviation and analyzed by two-way ANOVA, or mixed-effect analysis in cases of missing data, using Šidák correction for multiple comparisons (B,C,E,G-J), or by one-way ANOVA with pairwise comparisons of the upper and lower lobes, or of superficial and deep location of MSC IL-10 20 × 10 6 and 40 × 10 6 samples (D), or comparing other tissues to lung transplant (O).Results relative to GAPDH housekeeping gene (I,J).Quantitative PCR ng of human ALU DNA per 500 ng of total DNA, and * p < 0.0001 compared to all other tissues (O).Insets show magnified single-channel fluorescent images of the dashed area (K).BAL, bronchoalveolar lavage; EVLP, ex vivo lung perfusion; hIL-10, human interleukin-10; MSC, mesenchymal stromal cell; TX, transplantation.
A corresponding increase in lung hIL-10 was observed with tissue concentrations reaching 37 ± 16 and 76 ± 44 pg per mg of lung protein at the end of EVLP in the 20 × 10 6 and 40 × 10 6 MSC IL-10 groups, respectively (Figure 2C), far exceeding the levels achieved in previous adenoviral gene therapy studies [15].Similar hIL-10 levels were detected in different anatomical lung areas (Figure 2D), and an analysis of BAL samples indicated an increase in hIL-10 in the airway compartment (Figure 2E).

MSC IL-10 Therapy during EVLP Results in Transient hIL-10 Increase after Lung Transplantation
We used the pig single-lung transplant model to determine MSC IL-10 function and fate after transplantation.After finishing the 6 h EVLP, the left lung was transplanted into a recipient pig that was followed for 3 days (Figure 2F and Supplementary Table S1).MSC IL-10 treatment resulted in a dose-dependent early hIL-10 peak with recipient plasma levels reaching 261 ± 68 and 723 ± 276 pg/mL 2 h after transplant reperfusion in 20 × 10 6 and 40 × 10 6 MSC IL-10 groups, respectively (Figure 2G).Subsequently, recipient plasma hIL-10 levels declined to low levels within 1 to 2 days after transplantation and were undetectable at 3 days (Figure 2G).Similarly, lung tissue hIL-10 levels increased 1 h after transplant reperfusion but were not measurable from the transplant, or from the contralateral native lung, at 3 days (Figure 2H).BAL hIL-10 levels were also undetectable 3 days after transplantation (Figure 2E).
To confirm the presence of hIL-10 in the lungs at the mRNA level, lung samples were assessed with qRT-PCR using human-specific IL-10 primers.Mirroring the hIL-10 protein results (Figure 2C,H), lung hIL-10 mRNA levels were increased at the end of EVLP, and at 1 h after transplant reperfusion, and were undetectable 3 days after transplantation (Figure 2I).Also, human vascular endothelial growth factor (hVEGF) mRNA levels from the same lung samples were measured.Low, but detectable, hVEGF mRNA levels were observed (Figure 2J), suggesting that, in addition to the profoundly augmented hIL-10 production, MSCs IL-10 also secrete other paracrine factors, but at far lower levels.

MSCs IL-10 Delivered during EVLP Are Transiently Retained in the Lung
We next used human (HLA Class I and ALU)-and transgene (FLAG tag attached to the hIL-10 transgene)-specific markers, and multiplex fluorescent staining, to identify MSCs IL-10 in the lungs 6 h after EVLP start.HLA Class I + ALU + FLAG + cells were identified (Figure 2K), indicating that human cells with hIL-10 transgene expression were retained in the pig lung.MSCs IL-10 were relatively large (Figure 2K,L), resided in duplicate or as single cells (Figure 2M,N), and were localized mainly to the lung interstitium (Figure 2K,N) and occasionally to the alveolar space (Figure 2L,M).In addition, ALU PCR of lung samples and various recipient tissues was performed to detect human genomic DNA.ALU DNA was present in the lung at the end of EVLP, and was undetectable in the transplanted lung, contralateral native lung and all other non-transplant tissues of the recipient 3 days after transplantation (Figure 2O).To determine the effect of MSC IL-10 treatment on lung function, double lungs with 24 h cold ischemia were assigned to control, or pulmonary artery delivery of 20 × 10 6 or 40 × 10 6 MSCs IL-10 , in a blinded and randomized manner.The baseline 1 h parameters, obtained before MSC IL-10 administration, were similar in all groups (Supplementary Table S1).No significant differences were detected in pulmonary vascular resistance (Figure 3A), lung oxygenation (Figure 3B), peak airway pressure (Figure 3C) or dynamic compliance (Figure 3D) after the administration of 20 × 10 6 or 40 × 10 6 MSCs IL-10 .Additionally, the metabolic conditions in all groups remained stable during EVLP (Supplementary Figure S1), and no significant differences were found in markers for vascular permeability (Figure 3E-G).Analysis of lung apoptosis 6 h after EVLP start indicated that treatment with 20 × 10 6 MSCs IL-10 significantly decreased TUNEL + cell density in the lung alveolar compartment, whereas large airway epithelium apoptosis was infrequent in all groups (Figure 3H-K and Supplementary Figure S2).

MSC IL-10 Treatment Does Not Affect Lung Function after Transplantation
To evaluate the potential effects of MSCs IL-10 on post-transplant lung function, pulmonary parameters were collected after transplantation, and detailed analysis of transplanted graft oxygenation, ventilatory function, pulmonary artery pressures and histology were obtained immediately and 3 days after transplantation.MSC IL-10 treatment did not change early lung function, determined by selective graft upper and lower lobe P/F-ratios 1 h after reperfusion (Figure 4A), by the time to recipient extubation, or weaning from supplemental oxygen (Figure 4B).During the postoperative phase, no lung-related adverse events occurred, all recipients remained on room air and systemic P/F ratios remained stable (Figure 4C).Detailed lung evaluation 3 days after transplantation indicated no differences in transplant P/F ratios (Figure 4D), pulmonary artery pressures (Figure 4E) or ventilation parameters (Figure 4F,G).Additionally, histological examination indicated similar levels of acute lung injury in all groups (Figure 4H), and no differences in macroscopic lung appearance were detected between groups (Supplementary Figure S3).remained stable (Figure 4C).Detailed lung evaluation 3 days after transplantation indicated no differences in transplant P/F ratios (Figure 4D), pulmonary artery pressures (Fig- ure 4E) or ventilation parameters (Figure 4F,G).Additionally, histological examination indicated similar levels of acute lung injury in all groups (Figure 4H), and no differences in macroscopic lung appearance were detected between groups (Supplementary Figure S3).(n = 3) MSCs IL-10 during EVLP, were transplanted into recipient pigs with methylprednisolone and cyclosporine A background immunosuppression.(A) Early graft function was evaluated by measuring the P/F ratio from samples taken selectively from the upper or lower veins of the transplant 1 h after reperfusion.Postoperative recovery was assessed by the time from graft reperfusion to recipient (B) extubation or wean of supplemental oxygen, and by (C) determining the P/F ratio for up to 3 days after transplantation.Post-recovery P/F ratios were measured during spontaneous breathing of room air, and the 3-day values during anesthesia and mechanical ventilation.At 3 days after transplantation, (D) transplant upper and lower lobe P/F ratios were measured, and the transplant P/F ratio, (E) pulmonary artery pressures, (F) dynamic compliance and (G) tidal volume were assessed after elimination of the contralateral native lung blood flow and ventilation by clamping.(H) At 3 days, transplant acute lung injury score was determined by a transplant pathologist from histological sections assessing for air space hemorrhage, vascular congestion, edema/fibrin in the alveoli and the presence of infiltrating white blood cells.Each parameter was scored from 0 to 3, and a total acute lung injury score was calculated.

High-Dose MSC IL-10 Treatment Results in a Pro-Inflammatory Response
To explore possible immunological effects of the MSC IL-10 treatment, various inflammatory cytokines were measured.A modest gradual elevation in perfusate pro-inflammatory cytokines was seen in the control and 20 × 10 6 MSC IL-10 groups.In contrast, treatment with 40 × 10 6 MSCs IL-10 resulted in a significant increase in perfusate IL-1β (Figure 5A), IL-6 (Figure 5B), IL-8 (Figure 5C) and IL-10 levels (Figure 5D), measured by pig-specific cytokine ELISA.

Low-Dose MSC IL-10 Therapy Decreases Activated Macrophages in the Lung
As the inhibition of macrophage activation has been described as one of the mechanisms for the anti-inflammatory effects of IL-10 [27,28], we also analyzed th myeloid cell composition (Figure 7 and Supplementary Figure S5).No effects on lun trophils (Figure 7A), CD16 + monocytes (Figure 7B), CD163 + macrophages (Figure natural killer (NK) cells (Figure 7D) were found.However, macrophage subtype an revealed that treatment with 20 × 10 6 MSCs IL-10 decreased the number of DR + CD163 mid SSC low -activated macrophages (Figure 7F).

High-Dose MSC IL-10 Therapy Increases Immunity Early after Lung Transplantation
Finally, we performed mixed lymphocyte reaction assays to determine the o effect on early alloimmunity.MSC IL-10 treatment did not affect CD4 + cell proliferation ure 8A and Supplementary Figure S6).However, in contrast to 20 × 10 6 MSC IL-10 , trea with 40 × 10 6 MSC IL-10 increased CD8 + cell proliferation compared to the control (Figure 8B,C).Interestingly, this CD8 + cell proliferation occurred irrespective amount of stimulator cells present, suggesting that the immune activation related high MSC IL-10 dose was mainly alloantigen-independent.

High-Dose MSC IL-10 Therapy Increases Immunity Early after Lung Transplantation
Finally, we performed mixed lymphocyte reaction assays to determine the overall effect on early alloimmunity.MSC IL-10 treatment did not affect CD4 + cell proliferation (Figure 8A and Supplementary Figure S6).However, in contrast to 20 × 10 6 MSC IL-10 , treatment with 40 × 10 6 MSC IL-10 increased CD8 + cell proliferation compared to the control group (Figure 8B,C).Interestingly, this CD8 + cell proliferation occurred irrespective of the amount of stimulator cells present, suggesting that the immune activation related to the high MSC IL-10 dose was mainly alloantigen-independent.

Discussion
We combined cell and gene therapies and applied genetically engineered MSC with augmented hIL-10 production in translational large-animal EVLP and lung tr plant models.Administration of MSCs IL-10 in a clinically feasible way resulted in a ra efficient, and dose-dependent hIL-10 increase during EVLP and was followed by a tra ent hIL-10 elevation after lung transplantation.MSC IL-10 treatment did not affect perio ative lung function, but a comparison of two different MSC IL-10 doses showed diverg dose-related immunological effects, with the low MSC IL-10 dose leading to decreased ap tosis and lower macrophage activation, and the high MSC IL-10 dose resulting in profo pro-inflammatory effects and CD8 + cytotoxic T cell activation.
EVLP is an excellent platform for advanced therapeutics aiming for graft protect repair and immunomodulation [12].Here, we administered genetically modified MSC 10 during EVLP, and, in order to facilitate potential future translation to clinical trials, u pre-modified and cryopreserved MSCs IL-10 , clinical-grade EVLP and a large-animal tra plant model.Similar to previous studies, MSC administration through the pulmonary tery during EVLP resulted in MSC retainment in the lung [14,17,20].Additionally, present results provide important information about the subsequent fate of the MSC after lung transplantation, as they were still present 1 h after transplant reperfusion, were undetectable in the transplanted graft and all recipient tissues 3 days after operat

Discussion
We combined cell and gene therapies and applied genetically engineered MSCs IL-10 with augmented hIL-10 production in translational large-animal EVLP and lung transplant models.Administration of MSCs IL-10 in a clinically feasible way resulted in a rapid, efficient, and dose-dependent hIL-10 increase during EVLP and was followed by a transient hIL-10 elevation after lung transplantation.MSC IL-10 treatment did not affect perioperative lung function, but a comparison of two different MSC IL-10 doses showed divergent, doserelated immunological effects, with the low MSC IL-10 dose leading to decreased apoptosis and lower macrophage activation, and the high MSC IL-10 dose resulting in profound proinflammatory effects and CD8 + cytotoxic T cell activation.
EVLP is an excellent platform for advanced therapeutics aiming for graft protection, repair and immunomodulation [12].Here, we administered genetically modified MSCs IL-10 during EVLP, and, in order to facilitate potential future translation to clinical trials, used premodified and cryopreserved MSCs IL-10 , clinical-grade EVLP and a large-animal transplant model.Similar to previous studies, MSC administration through the pulmonary artery during EVLP resulted in MSC retainment in the lung [14,17,20].Additionally, our present results provide important information about the subsequent fate of the MSCs IL-10 after lung transplantation, as they were still present 1 h after transplant reperfusion, but were undetectable in the transplanted graft and all recipient tissues 3 days after operation.
Similarly, transgenic hIL-10 was elevated early after reperfusion, but decreased to low levels 1 to 2 days after transplantation.These results indicate that the viability of MSCs IL-10 after lung transplantation is limited.This is also supported by previous findings showing that live MSCs are detected in the lung up to 1 day after intravenous injection, but are then rapidly phagocytosed [29,30].Considering the kinetics of the transgenic hIL-10 in our study, therapy with genetically modified MSCs during EVLP would ideally target the early perioperative phase of lung transplantation, and could be thus potentially used to counteract ischemia-reperfusion injury and early alloimmune activation.
Importantly, we found that administration of 20 × 10 6 MSCs IL-10 during EVLP resulted in beneficial effects as it decreased lung apoptosis, lung IL-8 mRNA and macrophage activation.These results are in line with the generally known anti-inflammatory properties of IL-10 [28,31], and with the protective roles of IL-10 in lung ischemia-reperfusion [32,33] and transplantation [15,16].As alveolar macrophages are critical regulators of lung ischemiareperfusion injury [33][34][35], and as MSC IL-10 treatment increased hIL-10 in the alveolar space and decreased the number of activated macrophages, it is possible that alveolar macrophages mediated the beneficial effects of the 20 × 10 6 MSC IL-10 treatment.However, it is also possible that other IL-10 effects, and MSC effects independent of IL-10, contributed to the observed findings.First, in addition to controlling inflammation, IL-10 is also known to possess non-classical effects on epithelial proliferation and tissue repair [28].Second, besides the augmented hIL-10 production, conventional paracrine mediators secreted by the MSC IL-10 may have participated in the immunomodulation [17][18][19]-a possibility supported by our finding of human VEGF mRNA in the treated pig lungs.Third, as the viability of the MSC IL-10 after lung transplantation was limited, MSC apoptosis and phagocytosis may have affected the phenotype of antigen-presenting cells and resulted in changes in adaptive immunity [30,36].
Unexpectedly, the immunological effects of the high MSC IL-10 dose were significantly different, as treatment with 40 × 10 6 MSCs IL-10 resulted in the elevation of pro-inflammatory cytokines and the activation of CD8 + cytotoxic T cells.While this finding is contrary to the traditional anti-inflammatory characteristics of IL-10 [28,31], it is in line with the reported effects of high-dose IL-10 on CD8 + cytotoxic T cells [25,37].Specifically, several studies have demonstrated that high levels of IL-10 context-dependently induce IFN-γ and activate cytotoxic CD8 + T cells, resulting in the upregulation of perforin and granzymes [26,[38][39][40][41][42].These findings have important implications for anti-tumor immunity and, correspondingly, high doses of pegylated IL-10 have been used in preclinical experiments, and in clinical trials in cancer patients [25,26,40].As we found a profound upregulation of TNF-α, IFN-γ, perforin and granzyme B-prototype cytokines and mediators of cytotoxic cells [25,28]-in the transplanted lungs, and an increase in the number of IFNγ + CD8 + T cells, it is highly probable that the high-dose MSC IL-10 treatment activated cytotoxic T cells.Although the lung NK cell composition remained unchanged, cytotoxic effects on NK cells [43,44] and stimulatory effects on B cells could also have occurred [45].
Overall, our findings highlight the complex and divergent immunomodulatory effects of IL-10 and indicate that there is a therapeutic window for IL-10 in solid organ transplantation: low-to-moderate IL-10 levels have anti-inflammatory effects, probably achieved through macrophages, but very high levels may result in increased cytotoxicity after transplantation [46].When considering IL-10 therapy in solid organ transplantation, it is therefore essential to investigate dose-related effects, and to determine optimal IL-10 levels and kinetics after IL-10 administration.In our study, low-dose MSC IL-10 treatment increased perfusate IL-10 levels to 6 ng/mL and lung IL-10 levels to 37 pg/mg of lung protein during EVLP, while the respective IL-10 levels reached 17 ng/mL and 76 pg/mg of lung protein after high-dose MSC IL-10 treatment.Interestingly, in a previous study using human lungs, beneficial anti-inflammatory effects were achieved with adenoviral IL-10 therapy, resulting in lung IL-10 levels of approximately 30 pg/mg of lung protein [15], which is very close to the lung IL-10 levels in our low-dose MSC IL-10 group.On the other hand, treatment of cancer patients with daily subcutaneous PEGylated IL-10 resulted in im-

Figure 1 .
Figure 1.Study design and rapid IL-10 production of cryopreserved MSCs IL-10 in vitro.(A) Human umbilical cord MSCs were genetically engineered for augmented human IL-10 production (MSCs IL- 10 ) and cryopreserved.(B) In vitro bench testing of the cryopreserved MSC IL-10 batches for hIL-10 production (54 × 10 3 cells from the 20 × 10 6 MSC IL-10 group, n = 5; or from the 40 × 10 6 MSC IL-10 group, n = 3).(C) Pig double lungs subjected to 24 h cold ischemia were connected to EVLP for 6 h and randomized to control (n = 7), or pulmonary artery delivery of 20 × 10 6 (n = 5) or 40 × 10 6 cryopreserved MSCsIL-10 (n = 6).Subsequently, the left lung was transplanted into a recipient pig that received methylprednisolone and cyclosporine A background immunosuppression, and was sacrificed 3 days after transplantation.(D) A clinical-grade Toronto ex vivo lung perfusion system was used for the EVLP experiments, and MSCs IL-10 were administered to the EVLP circuit in a blinded and randomized manner with a syringe (white arrow) and a tubing extension was attached to the

Figure 1 .
Figure 1.Study design and rapid IL-10 production of cryopreserved MSCs IL-10 in vitro.(A) Human umbilical cord MSCs were genetically engineered for augmented human IL-10 production (MSCs IL-10 ) and cryopreserved.(B) In vitro bench testing of the cryopreserved MSC IL-10 batches for hIL-10 production (54 × 10 3 cells from the 20 × 10 6 MSC IL-10 group, n = 5; or from the 40 × 10 6 MSC IL-10 group, n = 3).(C) Pig double lungs subjected to 24 h cold ischemia were connected to EVLP for 6 h and randomized to control (n = 7), or pulmonary artery delivery of 20 × 10 6 (n = 5) or 40 × 10 6 cryopreserved MSCs IL-10 (n = 6).Subsequently, the left lung was transplanted into a recipient pig that received methylprednisolone and cyclosporine A background immunosuppression, and was sacrificed 3 days after transplantation.(D) A clinical-grade Toronto ex vivo lung perfusion system was used for the EVLP experiments, and MSCs IL-10 were administered to the EVLP circuit in a blinded and randomized manner with a syringe (white arrow) and a tubing extension was attached to the pulmonary artery cannula (black arrow).(E) Analysis of preformed hIL-10 in MSC IL-10 supernatant taken immediately prior to MSC IL-10 administration.Data mean ± standard deviation and analyzed with (B) two-way ANOVA using Šidák correction for multiple comparisons or (E) 2-tailed Student's t-test.Ad, adenovirus; EVLP, ex vivo lung prefusion; hIL-10, human interleukin-10; MSC, mesenchymal stromal cell; TX, transplantation.

2 .
MSC IL-10 Delivery during EVLP Results in Rapid hIL-10 Elevation in Perfusate, Lung Tissue and Airways

Figure 2 .
Figure 2. MSCIL-10 treatment during EVLP results in rapid dose-dependent hIL-10 elevation that decreases after lung transplantation.Pig double lungs subjected to 24 h cold ischemia were (A) connected to EVLP for 6 h, 20 × 10 6 (n = 5) or 40 × 10 6 (n = 6) MSCs IL-10 were administered, and hIL-10 was measured using ELISA.(B) After MSC IL-10 administration, EVLP perfusate hIL-10 levels were rapidly and dose-dependently elevated within minutes and continued to steadily increase for the duration of EVLP.(C) Lung tissue hIL-10 protein levels increased during EVLP, and (D) no significant differences were detected between samples taken from the upper or lower lobes, or from superficial or deep areas of the lung.(E) BAL hIL-10 was dose-dependently elevated 6 h after EVLP start.(F) After EVLP, the left lung was transplanted to a recipient pig that received methylprednisolone and cyclosporine A immunosuppression. (G) Recipient plasma hIL-10 protein levels peaked dose-dependently during the first 4 h after transplant reperfusion and declined to low levels within 1-2 days after transplantation.(H) Lung tissue hIL-10 protein levels were elevated 1 h after lung transplant reperfusion but were undetectable at 3 days.(I) Quantitative RT-PCR with human-specific primers of lung samples showed markedly increased hIL-10 mRNA expression at the end of EVLP, lower hIL-10 mRNA levels 1 h after transplant reperfusion and undetectable levels 3 days after transplantation.(J) Quantitative RT-PCR for human VEGF mRNA revealed low, but detectable levels of human VEGF mRNA in the lung 6 h after EVLP and 1 h after transplantation.(K) Lung tissue multiplex

18 Figure 3 .Figure 3 .
Figure 3. Lung function during EVLP after MSC IL-10 administration.Pig lungs with 24 h cold ischemia were connected to clinical-grade EVLP and randomized to control (n = 7), or to receive 20 × 10 6 (n = 5) or 40 × 10 6 MSCs IL-10 (n = 6) through the pulmonary artery 1 h after EVLP start.Lung function parameters were recorded every hour for 6 h.(A) Pulmonary vascular resistance, (B) delta pO2, (C) peak airway pressure and (D) dynamic compliance.Vascular permeability and lung edema were evaluated by (E) total perfusate loss and (F) lung weight gain during EVLP, and by (G) lung wet/dry ratio at the end of EVLP.Cellular apoptosis was evaluated from tissue sections at the end of EVLP by (H) TUNEL staining using automated image analysis and tissue segmentation to alveolar and large airway epithelium compartments.Representative TUNEL + staining of (I) control, (J) 20 × 10 6Figure 3. Lung function during EVLP after MSC IL-10 administration.Pig lungs with 24 h cold ischemia were connected to clinical-grade EVLP and randomized to control (n = 7), or to receive 20 × 10 6 (n = 5) Cells 2024, 13, x FOR PEER REVIEW 8 of 18

Figure 4 .
Figure 4. Effect of MSC IL-10 treatment on lung function up to 3 days after transplantation.Left single lungs, which had been randomly assigned to control (n = 5), or to receive 20 × 10 6 (n = 5) or 40 × 10 6 (n = 3) MSCs IL-10 during EVLP, were transplanted into recipient pigs with methylprednisolone and cyclosporine A background immunosuppression.(A) Early graft function was evaluated by measuring the P/F ratio from samples taken selectively from the upper or lower veins of the transplant 1 h after reperfusion.Postoperative recovery was assessed by the time from graft reperfusion to recipient (B) extubation or wean of supplemental oxygen, and by (C) determining the P/F ratio for up to 3 days after transplantation.Post-recovery P/F ratios were measured during spontaneous breath-

Figure 4 .
Figure 4. Effect of MSC IL-10 treatment on lung function up to 3 days after transplantation.Left single lungs, which had been randomly assigned to control (n = 5), or to receive 20 × 10 6 (n = 5) or 40 × 10 6

Figure 7 .
Figure 7. Lung transplant myeloid cell composition after MSC IL-10 treatment 3 days after tran tation.Myeloid cells and monocyte and macrophage subtypes in lung transplants 3 days after plantation were quantified by flow cytometry.Lung (A) neutrophils, (B) CD16 + monocyt CD163 + macrophages and (D) NK cells.(E) Monocytes were further divided according to SLA expression, and (F) macrophages according to CD163, side scatter and SLA II DR expression ment with 20 × 10 6 MSCs IL-10 significantly decreased the number of SLAII -DR + CD163 mid SSC l vated macrophages.Data mean ± standard deviation and analyzed by 1-way ANOVA with nett's correction comparing treatment groups to the control group.* p < 0.05.MSC, mesenc stromal cell; NK, natural killer; SLA, swine leukocyte antigen; SSC, side scatter.

Figure 7 .
Figure 7. Lung transplant myeloid cell composition after MSC IL-10 treatment 3 days after transplantation.Myeloid cells and monocyte and macrophage subtypes in lung transplants 3 days after transplantation were quantified by flow cytometry.Lung (A) neutrophils, (B) CD16 + monocytes, (C) CD163 + macrophages and (D) NK cells.(E) Monocytes were further divided according to SLA II DR expression, and (F) macrophages according to CD163, side scatter and SLA II DR expression.Treatment with 20 × 10 6 MSCs IL-10 significantly decreased the number of SLAII -DR + CD163 mid SSC lowactivated macrophages.Data mean ± standard deviation and analyzed by 1-way ANOVA with Dunnett's correction comparing treatment groups to the control group.* p < 0.05.MSC, mesenchymal stromal cell; NK, natural killer; SLA, swine leukocyte antigen; SSC, side scatter.

Figure 8 .
Figure 8.Effect of MSC IL-10 treatment on early alloimmune activation 3 days after transplanta Early alloimmune response was determined using mixed lymphocyte reaction.Donor spleen were used as stimulators, and recipient spleen cells, procured 3 days after transplantation, w used as responders.CD4 + cell (A) and CD8 + cell (B) proliferation with different stimulators to sponder cell ratios.Representative cell proliferation results of CD4 and CD8 cells with 1:1 stim tor: recipient cell ratio and eFluor 450 proliferation dye (C).Data mean ± standard deviation analyzed by 1-way ANOVA with Dunnett's correction comparing treatment groups to the con group.* p < 0.05, ** p < 0.01.MSC, mesenchymal stromal cell.

Figure 8 .
Figure 8.Effect of MSC IL-10 treatment on early alloimmune activation 3 days after transplantation.Early alloimmune response was determined using mixed lymphocyte reaction.Donor spleen cells were used as stimulators, and recipient spleen cells, procured 3 days after transplantation, were used as responders.CD4 + cell (A) and CD8 + cell (B) proliferation with different stimulators to responder cell ratios.Representative cell proliferation results of CD4 and CD8 cells with 1:1 stimulator: recipient cell ratio and eFluor 450 proliferation dye (C).Data mean ± standard deviation and analyzed by 1-way ANOVA with Dunnett's correction comparing treatment groups to the control group.* p < 0.05, ** p < 0.01.MSC, mesenchymal stromal cell.
Cells 2024, 13, x FOR PEER REVIEW 11 of 18 ddCt method using GAPDH as the housekeeping gene, and results are given in relation to respective pre-EVLP mRNA levels.Data mean ± standard deviation and analyzed by 2-way ANOVA (A-D), or by 1-way ANOVA (E-G), with Dunnett's correction comparing treatment groups to the control group.EVLP, ex vivo lung perfusion; IFN, interferon; IL, interleukin; MSC, mesenchymal stromal cell; TNF, tumor necrosis factor.