Human liver derived mesenchymal stromal cells ameliorate murine ischemia-induced inflammation through macrophage polarization

Introduction The immunomodulatory properties of mesenchymal stromal cells (MSC) have been well-characterized in in-vitro and in-vivo models. We have previously shown that liver MSC (L-MSC) are superior inhibitors of T-cell activation/proliferation, NK cell cytolytic function, and macrophage activation compared to adipose (A-MSC) and bone marrow MSC (BM-MSC) in-vitro. Method To test these observations in-vivo, we infused these types of MSC into mice with unilateral renal artery stenosis (RAS), an established model of kidney inflammation. Unilateral RAS was induced via laparotomy in 11-week-old, male 129-S1 mice under general anesthesia. Control mice had sham operations. Human L-MSC, AMSC, and BM-MSC (5x105 cells each) or PBS vehicle were injected intra-arterially 2 weeks after surgery. Kidney morphology was studied 2 weeks after infusion using micro-MRI imaging. Renal inflammation, apoptosis, fibrosis, and MSC retention were studied ex-vivo utilizing western blot, immunofluorescence, and immunohistological analyses. Results The stenotic kidney volume was smaller in all RAS mice, confirming significant injury, and was improved by infusion of all MSC types. All MSC-infused groups had lower levels of plasma renin and proteinuria compared to untreated RAS. Serum creatinine improved in micetreated with BM- and L-MSC. All types of MSC located to and were retained within the stenotic kidneys, but L-MSC retention was significantly higher than A- and BM-MSC. While all groups of MSC-treated mice displayed reduced overall inflammation and macrophage counts, L-MSC showed superior potency in-vivo at localizing to the site of inflammation and inducing M2 (reparative) macrophage polarization to reduce inflammatory changes. Discussion These in-vivo findings extend our in-vitro studies and suggest that L-MSC possess unique anti-inflammatory properties that may play a role in liver-induced tolerance and lend further support to their use as therapeutic agents for diseases with underlying inflammatory pathophysiology.


Introduction
Mesenchymal stromal cells (MSC) have been widely studied for their potential as therapeutic agents to treat a multitude of inflammatory pathologies due to their immunomodulatory capabilities.MSC have been derived from several types of tissues, but those isolated from adipose tissue and bone marrow are most often used in clinical trials.Guided by the liver's unique tolerogenic microenvironment and immunomodulatory properties, we postulate that liver-derived MSC (L-MSC) may have superior therapeutic potential.In fact, in-vitro studies that directly compared MSC isolated from healthy adult liver (L-MSC) to either those from adipose (A-MSC) or bone marrow (BM-MSC) demonstrate that L-MSC are superior at inhibiting the proliferation of alloreactive T cells, IFNy production by T cells (1), and the cytotoxic abilities of NK cells (2).Additionally, transcriptomic and proteomic analyses of A-, BM-, and L-MSC show significantly higher level of expression of several key immunomodulatory molecules in L-MSC (1).
Collectively, the in-vitro studies suggest that L-MSC possess a distinct genomic profile that may enhance their immunomodulatory capabilities compared to A-or BM-MSC.The goal of this study is to characterize the function of L-MSC in-vivo and evaluate if their superior immunomodulatory capabilities seen in-vitro translate into better function in-vivo.We examined the therapeutic and immunomodulatory function of L-MSC in the context of ischemic injury using the validated unilateral renal artery stenosis (RAS) mouse model and directly compared their effect to that of A-and BM-MSC.We hypothesized that L-MSC would be non-inferior in their ability to improve overall renal function in the stenotic kidney with greater influence on immunological changes compared to A-or BM-MSC.

Cell culture
The collection of MSC from healthy adults are approved by Mayo Clinic Institutional Review Board (IRB #17-007379 (liver), IRB #11-009182 (adipose tissue) and IRB # 10-002572 (bone marrow).All tissues are collected as part of scheduled donation procedures and informed consent are obtained prior to collecting tissue samples for this study.MSC are isolated and passaged from human adipose, bone marrow, and liver tissue as previously described (1)(2)(3).Specifically, adipose tissue is obtained from the subcutaneous compartment during the abdominal incision for a living donor nephrectomy procedure.Bone marrow aspiration from the iliac crest is performed by specialized hematology team under general anesthesia as part of living donor nephrectomy procedure.A liver biopsy sample, measuring 1cm x 1cm, is obtained from donor organs (deceased or living donor) for isolation of MSC.After obtaining tissue samples, the source tissue is enzymatically digested, and the plastic-adherent cells from the resulting cell suspension are placed into MSC culture media and are allowed to proliferate for 2 weeks before first passage.The cell lines used to date represent both sexes (50% female), racial heterogeneity (>10% non-Caucasians), and a wide range of ages from 20 to 75.Their phenotype and trilineage differentiation capacity were confirmed with flow cytometry and MSC functional identification assay (R&D Systems, Minneapolis, MN, USA), respectively.Prior to administration into mice, MSC (5x10 5 cells in 200ul PBS) in Passage 3 were fluorescently labeled with CellTrace ™ Far Red (CTFR, ThermoFisher Scientific, Waltham, MA, USA) to allow for detection after infusion.

Renal artery stenosis model
All protocols were approved by Mayo Clinic IRB and Institutional Animal Care and Use.As previously described (4), 11-week-old, male 129-S1 mice (Jackson laboratory, Bar Harbor, ME, USA) underwent open laparotomy under general anesthesia.After exposure of the right renal artery, a 0.15mm diameter arterial cuff was placed on the artery and secured with sutures to achieve partial occlusion of blood flow to the right kidney (i.e., stenotic kidney, STK).Two weeks following RAS surgery, fluorescently tagged MSC (5x10 5 cells in 200ul of PBS) derived from human adipose (A-MSC), bone marrow (BM-MSC), or liver (L-MSC) tissues, were given to RAS mice intra-arterially through direct cannulation of the carotid artery via vascular cut down.Mice that underwent surgery without cuff placement (n=4) served as negative controls (i.e.sham group).Mice that underwent RAS surgery but received an infusion of PBS (n=4) served as positive controls (i.e.untreated RAS group).Tail cuff blood pressures (Kent Scientific, Torrington, CT, USA) were also obtained at baseline, two weeks following RAS surgery, and two weeks following MSC infusion.General anesthesia was achieved using 3% isoflurane inhalation for induction and 1.5% during RAS surgery and intra-arterial MSC injection.Mice were euthanized after MRI imaging.Briefly, mice underwent general anesthesia with isoflurane as stated above.A midline abdominal incision (approximately 1-2cm in length) was made to access the peritoneal cavity.Peritoneal organs were then reflected superiorly to expose the inferior vena cava in order to obtain blood samples.After exsanguination, the STKs were collected for tissue processing.

Imaging protocol
Two weeks after MSC or PBS injection, mice were scanned using MRI as previously described (5).Previously established imaging protocols were used to acquire the appropriate images to quantify the volume, perfusion, and oxygenation of the STKs (5,6).All image analyses were performed using Analyze software (version 12.0; Biomedical Imaging Resource, Mayo Clinic, MN, USA) and Matlab (The MathWorks, Natick, MA, USA).

Serum and urinary biomarker measurements
Post MRI imaging, blood from the inferior vena cava and urine were collected at the time of euthanasia.Whole blood was centrifuged, and the resulting plasma was collected.Plasma renin concentration was measured by the Renin Assay Kit (Cat#MAK157, Millipore Sigma, St. Louis, MO, USA).Serum creatinine was measured using the Serum Creatinine Detection Kits (Cat# KB02-H, Arbor Assays, Ann Arbor, MI, USA).Urinary protein levels were measured using the Pierce ™ Bradford Protein Assay kit (Cat#23200, ThermoFisher, Waltham, MA, USA).All kits were used per manufacturer's instructions.

RT-PCR
Frozen STK samples were homogenized in 350ul of ice-cold lysis buffer, supplied by mirVana PARIS total RNA isolation kit (Cat# AM1556, ThermoFisher Scientific).Total RNAs were then isolated from homogenized samples according to the kit protocol.Total RNA concentrations were measured by a NanoDrop Spectrophotometer (NanoDrop).First-strand cDNA was produced from 800ng of total RNA using SuperScript VILO cDNA Synthesis kit (Cat#11755-050, ThermoFisher Scientific).Relative quantitative PCR were performed using Taqman assays, containing 4ul of cDNA products.All primers were purchased from ThermoFisher Scientific with the following catalog numbers: CD45 (Mm01293577); IFNy (Mm01168134); TNFa (Mm00443258); and GAPDH (Mm99999915).PCR analysis was done on Applied Biosystems Quantstudio 7 using the following conditions: 50°C for 2 minutes, 95°C for 10 minutes and 40 cycles of 95°C for 15 seconds and 60°C for 1 minute.Fold changes of gene expressions were calculated using 2-DDCT method.

Western blot
Frozen STK samples were homogenized, and protein expression was expressed by western blotting.Protein concentrations were measured using a BCA Protein Assay Kit (Cat# 23225, ThermoFisher Scientific) per manufacturer's instructions.The membranes were blocked with 5% BSA, incubated with primary antibodies, washed, and incubated with secondary antibodies at room temperature.Finally, the membranes were washed and incubated with ECL Western Blot Substrate (Cell Signaling Technology, Inc., Danvers, MA, USA) and were visualized on ImageQuant ™ LAS4000.Anti-IFNy (Cat# BS-0480R, Bioss, Woburn, MA, USA) and anti-TNFa (Cat# ab6671, Abcam, Waltham, MA, USA) antibodies were used as primary antibodies.GAPDH antibody was used to normalize the results.

Statistical analysis
All statistical analyses were performed using GraphPad Prism version 10.2.2 (324) for Windows (GraphPad Software, Boston, Massachusetts USA, www.graphpad.com).All data are expressed as either mean ± SD for normally distributed data or median [IQR] for non-normally distributed data.Hypothesis testing was carried out using one-way ANOVA followed by a student t-test for normally distributed data.Data not following normal distribution were analyzed using Kruskal-Wallis followed by Wilcoxon test.All data were considered significant if p<0.05.

Blood pressure
Initially, eight mice were randomly assigned to receive infusion of each type of MSC.At the conclusion of the study, two mice in the A-MSC group were lost due to total infarction of the STK, one mouse in the BM-MSC was lost due to hydronephrosis of the STK secondary to ureteral stricture, and one mouse in the L-MSC died just prior to MRI imaging, resulting in a final count of A-MSC (n=6), BM-MSC (n=7), and L-MSC (n=7) for analyses.Blood pressure using tail cuffs were obtained at baseline, post-RAS surgery, and post-MSC or PBS infusion.As expected, mean systolic (SBP) and diastolic blood pressure (DBP) measurements were higher than baseline after RAS surgery (Figure 1A).Injection of MSC did not show reduction of overall SBP or DBP nor in the amount of absolute or percent change in SBP or DBP from RAS surgery to post-MSC injection (data not shown).

Renal volume, perfusion, and oxygenation
Non-invasive evaluation of the volume, perfusion, and oxygenation of the STKs were performed using micro-MRI analysis.Compared to the sham group, untreated RAS mice had significant loss of volume in the STKs (94.18 ± 50.6mm 3 vs 266 ± 24.7mm 3 in sham, p< 0.001), suggestive of ischemic injury (Figures 2A, B).With MSC treatment, the volumes of the STKs significantly improved compared to the untreated RAS mice (A-MSC: 188.8 ± 17.6mm 3 ; BM-MSC: 226.1 ± 37.9mm 3 ; L-MSC: 181.9 ± 62mm 3 ; all vs RAS, p<0.001).No significant differences were noted in the volume of the STKs among the MSC treatment groups (Figure 2B).Cortical and medullary perfusion and oxygenation were also measured using micro-MRI.In this Representative MRI image of STK in coronal section (A).Non-invasive measurement of volume in the STKs (B) and the oxygenation and perfusion to the cortex and medulla in the STKs (C) within each group.All measurements are expressed as mean ± SD.For oxygenation, R 2 *(sec-1) reflects hypoxia with lower R 2 * indicating better oxygenation.STK, stenotic kidney.2C), but medullary perfusion in BM-MSC group was higher than in A-MSC group.
Representative images of inflammatory macrophages, M1 in (A) and reparative macrophages, M2 in (B) the STKs within each group.Manual counts of M1 and M2 per high power field (40x) and their ratio within the STKs are shown in (C) as median ± IQR.RAS, renal artery stenosis; A-MSC, Adipose mesenchymal stromal cells; BM-MSC, Bone Marrow mesenchymal stromal cells; L-MSC, Liver mesenchymal stromal cells.

Discussion
In this study, we aimed to characterize the effect of the novel L-MSC in-vivo and to directly compare their impact on ischemic injury to the more established A-and BM-MSC.We demonstrated that L-MSC are equally as effective as A-MSC and BM-MSC at improving renal function, the volume, and oxygenation of the renal medulla in the STKs.Additionally, L-MSC-treated RAS mice achieved a similar reduction in inflammation in the STKs as those treated with A-and BM-MSC.However, significantly more L-MSC were retained in the STKs, and L-MSC-treated mice had greater polarization of macrophages toward a more reparative (M2) phenotype compared to A-and BM-MSC treated groups.
MSC have been extensively investigated as therapeutic agents for inflammatory conditions, classically in graft versus host disease and inflammatory bowel disease, but also in ischemic renal injury (9)(10)(11)(12).Although the clinical efficacy of MSC treatments has been variable, data from experimental and human clinical trials support MSCs' immunomodulatory potential through intricate communications with both the innate and adaptive immune system via several proposed routes, including paracrine secretions, direct cell-to-cell contact, and release of exosomes.The downstream effect is the resolution of inflammation and the promotion of tissue regeneration through several mediatory pathways such as induction of M2 macrophage polarization (12, 13).
The source tissue of MSC and the microenvironment in which they are found impact MSC functions and properties.The liver is often considered to be an immunologically privileged organ that serves as a critical immune interface (14).Several clinical studies involving simultaneous liver and kidney transplant or simultaneous liver and heart transplant demonstrate that compared to solitary kidney or heart transplants, the presence of concomitant liver allograft was protective against both T cell and antibody-mediated rejection and overall improved graft survival (15)(16)(17)(18).On a cellular level, simultaneous liver and kidney transplant recipients demonstrated lower frequency of circulating CD8 + , activated CD4 + , and effector memory T cells and had decreased alloreactivity to donor cells compared to solitary kidney transplant recipients (16).Likewise, secretome analysis of simultaneous liver and kidney transplant recipients showed downregulation of inflammatory pathways and upregulation of tissue integrity pathways (17).Taken together, the superior immunomodulatory properties of the MSC isolated from liver may be closely associated with the immune context surrounding the organ.
Our findings in this study underscore previous studies that demonstrated improvement in renal function, oxidative stress, and inflammation after MSC treatment (4,19,20) as well as the impact of MSC on macrophage polarization (12).However, the current study augments the previous bodies of literature in several ways.We directly determined and compared the positive impact of MSC isolated from liver tissue, which has not been explored in detail to our best knowledge as a therapeutic agent, to that of more established MSC isolated from adipose and bone marrow tissues.Additionally, we demonstrated that a significantly higher number of L-MSC homed to site of injury than A-MSC and BM-MSC and exhibited greater impact on macrophage phenotypes.Interestingly, for more structural related changes, only A-MSC treated group achieved reduction in fibrosis while BM-and L-MSC treated groups showed significant reduction in levels of reactive oxygen species.While MSCs generally share many similar characteristics, previous studies have demonstrated significant differences among A-, BM, and L-MSCs that may explain some of the differing effects we observed in this study.For example, in-vitro studies have shown that L-MSC have a more homogenous migration kinetics toward chemoattractants than A-MSC, while the latter have superior antifibrotic and pro-angiogenic properties (21-24).Macrophage polarization plays a major role in liver disease (25).M1 macrophages promote tissue injury in vast majority of the liver diseases (viral, alcohol-related and metabolic-associated), whereas M2 macrophages attenuate liver injury and inflammation (26).At steady state, the liver microenvironment favors M2 polarization (27) for homeostasis.Interestingly, here, we demonstrate that adoptive transfer of human L-MSC in a mouse model of inflammation also promotes M2 polarization.Thus, it is possible that L-MSC have a role in liver homeostasis, which will need to be investigated further in the future.
Our study is not without limitations.The MSC treated groups did not result in improvement in blood pressures and perfusion or decrease in apoptosis compared to untreated RAS group.Given our small sample size, it is possible that our study may not have been adequately powered to evaluate all these physiological and histological changes.We also found that despite L-MSC-treated RAS mice having lower plasma renin, the urinary protein level was higher compared to the A-MSC group.This might be due to differential impact of MSC types on cells in the juxtaglomerular apparatus.Additionally, we noted discordance between IFNy gene expression and protein expression for A-and BM-MSC treated mice.The elevated IFNy protein expression in the A-and BM-MSC groups, but not in L-MSC group, could be related to post-transcriptional regulation.Indeed, previous transcriptomic analysis comparing A-, BM-, and L-MSC demonstrated significant upregulation of INFy regulatory genes in L-MSC (1,2), further supporting that L-MSC likely exert greater influence on the immune system than A-or BM-MSC.Additionally, while some of the superior effects on macrophages might have resulted from the engraftment of a larger number of L-MSC compared to A-and BM-MSC, such differences were not consistently observed in other parameters.Therefore, cell number may not have been the sole determinant of L-MSCs' effects.In our study, mice were also given a single infusion of MSC.Multiple infusions may be needed in order for MSC to exert maximal effect on the ischemic injury to the kidney (28).Additionally, more time than the allocated two weeks in this study may have been needed to see a more pronounced impact of reduced inflammation on renal function in the MSC-treated groups.
In summary, our study established the effect of L-MSC in-vivo on ischemic injury and directly compared their impact to that of A-MSC and BM-MSC.We showed that L-MSC are as effective as the commonly studied A-and BM-MSC at mitigating ischemic renal injuries.Furthermore, they are superior at homing to site of injury and at inducing polarization toward reparative macrophages when compared to A-and BM-MSC.Based on these findings, we are currently exploring the effect of local delivery of MSC on alloimmune mediated damages through direct infusion into the allograft renal artery in our ongoing clinical trials with adult renal transplant recipients (NCT05456243).As part of the clinical trial, we are collaborating with the Mayo Clinic Center for Regenerative Biotherapeutics Laboratory (IRB 17-007379) to routinely generate and culture MSC cell lines from adipose, bone marrow, and liver tissue (1cm x 1cm biopsy sample) from healthy adult donors in a GMP facility and testing for MSC phenotypic markers and trilineage differentiation to meet the release criteria for clinical use.More work will need to be done to detail the mechanism(s) through which L-MSC interact with the immune system to effectuate their impact on the surrounding environment.

1
FIGURE 1Blood pressure (mean ± SD) measured by tail cuff within each group at baseline, after RAS surgery, and after PBS or MSC infusion are shown in (A).Serum creatinine, urinary protein, and plasma renin levels for each group are shown in (B).All levels are expressed as mean ± SD.RAS, renal artery stenosis.