Autologous Mesenchymal Stem Cell Therapy for Idiopathic Pulmonary Fibrosis and Comprehensive Assessment of Circulating Immune Populations Cells


 BackgroundIdiopathic pulmonary fibrosis (IPF) is a chronic, progressive pulmonary disease characterized by aberrant tissue remodeling, formation of scar tissue within the lungs and continuous loss of lung function. The areas of fibrosis seen in lungs of IPF patients share many features with normal aging lung with cellular senescence being one. The contribution of the immune system to the etiology of IPF remains poorly understood. Evidence obtained from animal models and human studies suggests that innate and adaptive immune processes can orchestrate existing fibrotic responses. Currently, there is only modestly effective pharmacotherapy for IPF. Mesenchymal stem cells (MSCs)-based therapies have emerged as a potential option treatment of IPF. This study explores the possibility of using autologous MSCs as an IPF therapy and present an attempt to determine if the disease occurring in the lungs can be reflected on peripheral blood immune status. MethodsComprehensive characterization of autologous adipose derived MSCs (aMSCs) from five IPF patients and five age and gender matched Healthy Controls (HC) was done using flow cytometry, Droplet Digital PCR (ddPCR), Multiplex Luminex xMAP technology and confocal microscopy. For assessing the self renewal capacity and osteogenic differentiation IncuCyte Live Cell Imaging technology was used. Multi-parameter quantitative flow cytometry of un-manipulated whole blood of another group of 15 IPF patients and 87 (30 age and gender matched) HC was used to analyze 110 peripheral phenotypes.ResultsThere are no differences between autologous aMSCs from IPF patients and HC in their stem cell properties, self renewal capacity, plasticity for osteogenic differentiation, secretome content, cell cycle inhibitor marker levels and mitochondrial health. IPF patients had altered peripheral blood immunophenotype including reduced B cells subsets, increased T cell subsets, and increased granulocytes among others demonstrating clear and significant differences.ConclusionsOur results indicate that there is no difference in aMSCs properties from IPF patients and HC, suggesting that autologous aMSCs may be an acceptable option for IPF therapy.Characterization of peripheral immune phenotype may be a valuable indicator for successful therapy, and for potentially staging the disease.


Abstract Background
Idiopathic pulmonary brosis (IPF) is a chronic, progressive pulmonary disease characterized by aberrant tissue remodeling, formation of scar tissue within the lungs and continuous loss of lung function. The areas of brosis seen in lungs of IPF patients share many features with normal aging lung with cellular senescence being one. The contribution of the immune system to the etiology of IPF remains poorly understood. Evidence obtained from animal models and human studies suggests that innate and adaptive immune processes can orchestrate existing brotic responses. Currently, there is only modestly effective pharmacotherapy for IPF. Mesenchymal stem cells (MSCs)-based therapies have emerged as a potential option treatment of IPF. This study explores the possibility of using autologous MSCs as an IPF therapy and present an attempt to determine if the disease occurring in the lungs can be re ected on peripheral blood immune status.

Methods
Comprehensive characterization of autologous adipose derived MSCs (aMSCs) from ve IPF patients and ve age and gender matched Healthy Controls (HC) was done using ow cytometry, Droplet Digital PCR (ddPCR), Multiplex Luminex xMAP technology and confocal microscopy. For assessing the self renewal capacity and osteogenic differentiation IncuCyte Live Cell Imaging technology was used.
Multi-parameter quantitative ow cytometry of un-manipulated whole blood of another group of 15 IPF patients and 87 (30 age and gender matched) HC was used to analyze 110 peripheral phenotypes.

Results
There are no differences between autologous aMSCs from IPF patients and HC in their stem cell properties, self renewal capacity, plasticity for osteogenic differentiation, secretome content, cell cycle inhibitor marker levels and mitochondrial health. IPF patients had altered peripheral blood immunophenotype including reduced B cells subsets, increased T cell subsets, and increased granulocytes among others demonstrating clear and signi cant differences.

Conclusions
Our results indicate that there is no difference in aMSCs properties from IPF patients and HC, suggesting that autologous aMSCs may be an acceptable option for IPF therapy.
Characterization of peripheral immune phenotype may be a valuable indicator for successful therapy, and for potentially staging the disease.
Background Page 3/28 Idiopathic Pulmonary Fibrosis (IPF) is a complex disorder caused by multiple injuries to lung epithelium which triggers a local immune response leading to dysregulation of cellular homoeostasis [1].
Accumulations of extracellular matrix and scar formation in IPF are consequences of impaired wound repair mechanisms [2,3,4]. The pathogenesis is still poorly understood, and with the exception of lung transplantation, currently there are no signi cantly effective pharmacotherapies for IPF [2,[5][6][7][8]. Growing bodies of evidence from basic science and translational research indicate that IPF appears to be a direct result of immune dysregulation and aberrant wound healing response in the lungs [2,9,10]. Not much is known about the contribution of the immune system to the development of IPF or how lung immune responses affect the systemic immunity. To the best of our knowledge, there are no studies which comprehensively investigate the immune status of IPF patients, although there are a few studies identifying potential leukocytes involved with IPF pathogenesis [8, [11][12][13][14][15].
MSCs are important regulators of tissue repair and wound healing processes, have anti-in ammatory properties and display signi cant immunomodulatory capacity [16]. Use of MSCs-based therapy has emerged as a potential option for treatment of IPF. We have developed an expansive clinical program [17][18][19][20][21]  Abdominal wall adipose tissue (approximately 1.5 -2.5 g) was obtained under sterile conditions from IPF patients and age and gender matched healthy donors in an outpatient surgical suite. Tissues were processed with two hours of procurement. Cells were expended ex vivo according to the protocol based on Standard Operation Procedures for isolation, extraction and expansion of aMSCs analogs for clinical use [30,31]. In brief, after micro dissection the fat tissue was digested with collagenase Type I at 0.075% w/v, (Worthington Biochemicals, Lakewood, NJ) for 1.5 h at 37°C. Adipocytes were separated from the vascular fraction by centrifugation (400 x g for 5 min, at room temperature). The cell pellet was washed with PBS and passed through cell strainers (70μm followed by 40μm, BD Biosciences, Franklin Lakes, New Jersey). The resulting cell fraction was plated in T-75 cm 2 asks (Thermo Fisher, Waltham, MA) and incubated in a fully humidi ed incubator supplied with 5%CO 2 in PLGold xeno-free media. The xeno-free media named "PLGold media" consists of: Advanced MEM (Thermo Fisher Scienti c) supplemented with 5% (v/v) PLTGold, (Mill Creek Life Sciences, Rochester, MN), 1% (v/v) GlutaMax (Thermo Fisher) and 1% (v/v) antibiotics (100 U/ml penicillin, 100 g/ml streptomycin, HyClone, Logan, UT). Cells were propagated when they were 60-80% con uent using TrypLE (Trypsin-like Enzyme, Invitrogen, Carlsbad, CA) [30]. Cell yield and viability was quanti ed using Acridine Orange (AO) and Propidium Iodide (PI) nuclear stains for exclusion assay on a Luna-FL Dual Fluorescence Cell Counter (all from Logos Biosystems, Annandale, VA). All aMSCs used in the experimental procedures were between passages two and ve.
The proliferation and growth rate of aMSCs, was monitored by adding IncuCyte NucLight Rapid Red Reagent for Nuclear Labeling at 1:500 dilution (Essen Bioscience, Ann Arbor, MI) to the media. After a 30minute incubation at 37 0 C in a fully humidi ed incubator supplied with 5% CO 2 , it was placed in the IncuCyte S3 Live Cell Analysis instrument (Sartorius, Ann Arbor, MI) for uorescent quanti cation of cell proliferation. Fluorescent images of red nuclei from sixteen elds in each well were captured at 681nm every six hours with 10x objective. Each cell count was repeated twice in four replicas. The data acquisition, visualization and analysis were done using internal IncuCyte S3 Analyzing Software. The growth rate kinetic and doubling times (t d ) were evaluated recording the cell proliferation rate by counting A laser scanning confocal microscope was used to collect 2D and 3D cell images LSM 780 and ZEN 2010 software (Carl Zeiss, NY After 30 min incubation at 37 0 C in a fully humidi ed incubator supplied with 5% CO 2 plates were placed in an IncuCyte S3 Live Cell Analysis instrument for red nuclei count and green lipid droplets total green integrated intensity (GCU) imaging using 20 x objective. Fluorescent images of red nuclei (imaged at 681 nm) and green lipid GCU (imaged at 585 nm) from 16 elds in each well were captured every six hours for 24 hours. The data acquisition, visualization and analysis were done using internal IncuCyte S3 Analyzing Software. Each GCU value per well was normalized to the number of cells (red nuclei count) per well.
Secretome Analysis of Resting aMSCs aMSCs were seeded at 2105 cell/cm 2 in 6 well pates with 2.5 ml PLGold media/well. After 48 hours, the media was replaced with 2 ml fresh media/well. One well containing media only was used as a control for media content of analytes, which values were used as a background in secretome analysis.
After four days (96 h), media was collected, spun for ve minutes at 750 x g and supernatants were stored at -20 0 C until use. Immediately after collecting the media, 1 ml of fresh media containing 1:500 diluted IncuCyte NucLight Rapid Red Reagent was added to the cells. Cell number (red nuclei count) was counted in an IncuCyte S3 Live Cell Analysis instrument. For determining aMSCs secretome content, a 20plex custom made kit (Human Cytokine/Chemokine, Human Bone and Adipokine Magnetic bead panel, EMD Millipore, Burlington, MA), and Luminex xMAP technology (R&D Systems Inc., Minneapolis, MN) were used. The secretome assay was done in triplicate and was performed following the manufacturer's instructions. The plates were read by the MAGPIX instrument using xPONENT software for acquisition (Luminex, Austin, TX). Data analysis of the Median Fluorescence Intensity (MFI) and Coe cient of Variance (CV%) estimation were done by MILLIPLEX Analyst 5.1 software (EMD Millipore). The analyte concentrations (pg/ml) were normalized to 1x10 6 cells.

Analysis of aMSCs Senescence Status by Droplet Digital Polymerase Chain Reaction (ddPCR)
For establishing the senescence status of aMSCs we developed a Droplet Digital PCR (ddPCR) protocol for estimating the transcription level of CDKN1, p16 INK4A , p53, and RB1 cell cycle inhibitor markers. Total RNA from 1.5 x 10 6 cell pellets was extracted using RNeasy Mini Kit (Qiagen, Germantown, MD). The reverse transcription reaction was performed with random, Oligo(dT) 20  For the other tested cell cycle inhibitor markers as well as for TATA Binding Protein (TBP) as reference gene, commercially available uorescent labeled expression primers and probes were used (Bio-Rad).
The ddPCR reaction setup was as previously described [34] .The nal concentration of primers and probes in the reactions were 900 nmol/L and 250 nmol/L, respectively. Multiwall plates were sealed, vortexed brie y, centrifuged and placed on an automated droplet generator (AutoDG-Bio-Rad). Each sample was partitioned into 15,000-20,000 droplets. PCR ampli cation was performed on a Veriti Thermal Cycler (Applied Biosystems). The initial heating at 95°C for 10 minutes was followed by 60 cycles of denaturation at 94°C for 30 seconds, annealing and extension at 58°C for 1 minute, and a nal extension step at 98°C for 10 minutes. The completed reactions were stored at 4°C until reading them on a QX200 droplet reader (Bio-Rad). Data analysis was performed using 2D Module of the QuantaSoft software (BioRad).
Quantitative Flow Cytometry Assay for Immunopro ling

Procedural Factors
A separate group of 15 IPF patients were identi ed from the Interstitial Lung Diseases Outpatient Clinic by an expert pulmonologist in IPF and other brotic diseases of the lung following the ATS/ERS/JRS/ALAT Statement criteria [29]. All aspects of this study involving samples from IPF patients and age and gender matched healthy volunteers were reviewed and approved by the Mayo Clinic Institutional Review Board. All subjects provided written informed consent to participate.
Data are presented as mean ± SD, FVC% predictive = % of Forced Vital Capacity; FEV1% predictive = % of Forced Expiratory Volume in the 1st second, predictive; VC max % predictive = % Maximal Vital capacity.
"Predictive" means values adjusted for patient age, gender, and race.
To characterize the circulating immune phenotype, peripheral blood samples from 87 healthy volunteers (30 of which age and gender matched) and from a separate group of 15 IPF patients were collected in K 2 EDTA tubes (Becton Dickinson, Franklin Lakes, NJ) at initial or return visits. Un-manipulated whole blood was stained with antibodies directly within 12 hours of collection. Appropriate antibodies, undiluted (vendors, catalog numbers and amounts added per sample are listed in Additional File 1 and in [35, 36]) were added directly to the blood samples. Quantitative ow cytometry was performed to comprehensively assess 110 leukocyte populations and phenotypes from lymphocytes, monocytes, and granulocytes. All 10-color procedures, antibodies, ow protocols, instrument settings, and gating strategies for peripheral blood ow cytometry have been previously described by Gustafson et al. [35,36]. The ow cytometry data were analyzed using Kaluza 2.1 software (Beckman Coulter), allowing quanti cation of the absolute number as well as percent of immune cell subtypes. Fluorescently labeled isotypes were used as a control for each tested cell line.

Statistical Analysis
Results are expressed as mean ± SD. Statistical analysis was performed using GraphPad Prism 8 software. Intergroup comparisons of parametrically distributed continuous data were done using unpaired two-tailed Student's ttest. Differences were considered signi cant when p values *p<0.05, ** p<0.01, *** p<0.001, **** p<0.0001. Correlations between IPF patients' Pulmonary Function Test (PFT) values were established by calculating the Pearson correlation coe cient (r). Flow cytometry data are either represented as percentage of population or number of cells/ml. ddPCR data and presented as mean of three with CV%.

Discussion
In this study we explore the possibility of using autologous adipose MSCs as an IPF therapy. MSC's mode of action is still not fully understood, but they target the sites of injury, inhibit in ammation, and contribute to epithelial tissue repair [17-21, 24, 41]. Promising results of preclinical studies using MSCs suggest that they may represent a potential therapeutic option for the treatment of chronic lung diseases including IPF [22][23][24]. Subcutaneous adipose tissues MSCs are frequently used for clinical applications because of their easy access, minimally invasive procedure and their therapeutic potentials having been extensively studied [30,42].
This study examines the biological phenotype of adipose aMSCs isolated from IPF patients (IPFaMSCs) and age and gender matched healthy controls (HCaMSCs) expanded in PLGold xeno-free media to establish their suitability for IPF treatment.
Comprehensive characterization of aMSCs from both groups was carried out using a combination of ow cytometry, ddPCR, confocal microscopy, IncuCyte Live Cell Imaging and LUMINEX xMAP technologies. Our data provides evidence that autologous adipose tissue derived MSCs from IPF patients exhibit the same properties as the adipose tissue derived MSCs from the HC. (i) the aMSCs were adherent to plastic and exhibited the same small, spindle shape morphology, (ii) all tested cell lines were more than 99% positive for typical MSCs markers, and negative for lineage markers expression [32, 37, 38], (iii) the growth curves of IPFaMSCs and HCaMSCs cell lines display three distinct phases: initial short lag phase, followed by exponential growth and nally the plateau phase which is in agreement with the general behavior of the MSCs [43]. All cell lines have comparable exponential growth and have the same doubling times, which indicates identical potency for self-renewal, (iv) all aMSCs display similar health status, evaluated by analyzing the cells and mitochondrial morphology and assessing mitochondrial volumes.
Changes in mitochondrial volume have been associated with a wide range of important biological functions and pathologies [44]. Current evidence suggests that the areas of brosis seen in IPF patients' lungs share many mitochondrial dysfunction features [45]. The mitochondria of both tested cell lines have a typical shape and distribution for healthy MSCs mitochondria [46]. The IPFaMSCs mitochondrial volume is not different from the mitochondrial volume of HCaMSCs, which indicates that IPFaMSCs are healthy, not in a state of stress and are fully functional, (v) the capacity and plasticity for adipogenic differentiation of IPFaMSCs was the same as of HCaMSCs, (vi) MSCs are known to interact and actively communicate with their surrounding microenvironment through the secretion of cytokines and growth factors. Both cell groups secrete the same amount of growth factors, anti-in ammatory as well as proin ammatory cytokines, which indicates that the cell lines from both groups have the same capacity for regulating the tissue regeneration, proliferation, angiogenesis and modulation of in ammation, (vii) it has been long recognized that cellular senescence signi cantly contributes to the aging-related declines in tissue regeneration capacity and in the pathogenesis of aging-related diseases, such as IPF. The mechanism of how senescent cells contribute to aging and aging-related diseases remains unclear. Resident stem cells are particularly sensitive to senescence stresses. One of the hypotheses is that cellular senescence leads to exhaustion of the resident stem cells, which, in turn, causes a decline in tissue regenerative capacity during aging or upon injury [47].
The ndings in our study are that IPFaMSCs have the same expression pro le of cell cycle marker inhibitors as HCaMSCs and that there are no present senescence features in IPFaMSCs caused by the disease.
While establishing the suitability of using autologous aMSCs for IPF treatment, we also wanted to nd out whether an organ disease, as it is in the case of lungs with IPF, is re ected on the systemic immunity. To do so, we assessed the immune status of IPF patients by quantifying their circulating phenotypes and compared them with healthy controls. In addition, we sought to identify changes in IPF patients' immunological phenotypes, which correlate with their lung function indices.
Circulating granulocytes, neutrophils and eosinophils were found elevated in IPF patients' blood. These cells are rst to respond to the presence of pathogens in human lungs or upon tissue damage. They migrate from periphery to the damaged lungs in response to secreted chemokine and interleukin signals from invaded lungs and become activated [48,49]. They were found elevated in bronchoalveolar lavage uid (BALF) and sputum of IPF patients [50], as well as in peripheral blood, sputum and BALF in chronic obstructive pulmonary disease (COPD) patients [11]. Granulocytes and neutrophil counts in IPF patients in our study inversely correlate with all three measured pulmonary indices. High levels of neutrophil elastase were found in lung parenchyma and in both BALF and IPF patient serum. Therefore, neutrophils might indeed play an important role in the pathogenesis of IPF [51]. The same result was found in IPF patients BALF along with the increased IL-8 concentrations [50], and it was speculated that those ndings might be predictive for future exacerbations of IPF [52]. There is evidence that neutrophils might promote brosis via their regulation of extra cellular matrix (ECM) turnover [51,53,54]. Kinder et al. reports that an increased number of neutrophils in BALF is associated with early mortality in IPF [55]. All of these effects/interactions are complex and multifaceted and are not fully studied or understood in the case of IPF.
Natural Killer (NK) cells are the most important cell subsets involved in the production of IFN-γ , a cytokine suggested to improve survival of IPF patients [56,57]. In our study, only the percentage of [CD56

+ CD16-] NK cells in IPF patients' blood is elevated in comparison with HCs. A high percentage of this subset of NK cells was also observed in the blood of bone marrow transplant patients [58]. The functional study of isolated [CD56 + CD16-] cells indicates that these cells have very low cellular toxicity
[58]. The dynamic nature of cytokine and cellular pro le of the microenvironment in uences the development of speci c NK subtypes which may lead to conversion from pro-in ammatory to proresolution NK subtypes [59]. If there is any defect or disturbance in this process, it may lead to more severe in ammation and eventually to airway damage which can be re ected in the circulating phenotypes [12].
Recent studies have shown that B cells, as part of adaptive immunity, are involved in IPF pathology [13,60]. In our study there is no difference in the total number of B cells and the percentage of Plasma B cells between two tested groups. However, the Transitional B cells were lower in IPF patients than in HCs. As immature B cells emigrate from the bone marrow and enter the blood stream, they migrate toward the wounded organ (the lungs in the case of IPF) being attracted by secreted chemokines from the wounded lungs [13]. This can explain our ndings of decreased circulating Transitional B cell percentage in IPF patients' blood. At the same time our measurements of circulating [CD27 + IgM-IgD-] cell percentage shows an increase in IPF patients. It has been found that stimulated [CD27 + IgM-IgD-] cells activate telomerase and are responsible for age-related exhaustion of the B cells [14,61,62]. Since immune exhaustion and telomerase dysfunction have been implicated in IPF pathology, this subset of B-cells may be a good topic for further research.  [75].
Monocytes are known to contribute to the pathogenesis of idiopathic pulmonary brosis as was shown in the retrospective, multicenter cohort study [76]. The robust association of high monocyte count associated with mortality in other brotic diseases suggests they might contribute to the pathogenesis of these diseases as well. In our study, IPF patients have a higher percentage of circulating [CD33+] monocytes, which may contribute to the progression of the disease.

Conclusion
MSCs-based therapies for IPF, so far, are using allogeneic cells for treatment. Although the safety and tolerance of their use was con rmed, there are still minor side effects present, which might be circumvented by using autologous MSCs. Our study shows that adipose MSCs from IPF patients are not part of IPF pathology and may be used for IPF therapy.
Understanding the immunological status of IPF patients may provide insight into their immunity and its role in etiology of the disease. To our knowledge, our study is the most comprehensive evaluation of circulating phenotype in IPF patients so far and provides further evidence for the role of adaptive immunity in the pathogenesis of IPF. The increased Tregs in the IPF patients' peripheral blood correlate inversely with disease severity. Treg subpopulations may be promising prognostic factors for IPF. Characterization of the peripheral immune phenotypes in IPF patients may answer the question whether or not the immune events identi ed in the circulation can be used: as a monitor for personalized IPF therapy, to potentially classify/stage the disease, and identify those more likely to respond to therapy. Additional studies are needed with an expanded cohort of patients for positive identi cation of circulating phenotypes from peripheral blood as potential biomarkers for IPF. The funders did not play any role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.