Specific features of ex‐obese patients significantly influence the functional cell properties of adipose‐derived stromal cells

Abstract Adipose‐derived stromal cells (ADSC) are increasingly used in clinical applications due to their regenerative capabilities. However, ADSC therapies show variable results. This study analysed the effects of specific factors of ex‐obese patients on ADSC functions. ADSC were harvested from abdominal tissues (N = 20) after massive weight loss. Patients were grouped according to age, sex, current and maximum body mass index (BMI), BMI difference, weight loss method, smoking and infection at the surgical site. ADSC surface markers, viability, migration, transmigration, sprouting, differentiation potential, cytokine secretion, telomere length and mtDNA copy number were analysed. All ADSC expressed CD73, CD90, CD105, while functional properties differed significantly among patients. A high BMI difference due to massive weight loss was negatively correlated with ADSC proliferation, migration and transmigration, while age, sex or weight loss method had a smaller effect. ADSC from female and younger donors and individuals after weight loss by increase of exercise and diet change had a higher activity. Telomere length, mtDNA copy number, differentiation potential and the secretome did not correlate with patient factors or cell function. Therefore, we suggest that factors such as age, sex, increase of exercise and especially weight loss should be considered for patient selection and planning of regenerative therapies.


| INTRODUC TI ON
In recent years, the transplantation of autologous mesenchymal stromal cells (MSC) has played an increasingly important role in multiple regenerative therapies, including the treatment of various diseases and tissue injuries. 1 MSC are stromal progenitor cells with proliferation potential and the capability for multilineage differentiation as well as immunomodulation and considered to be a promising treatment modality for tissue regeneration and repair ( Figure 1). MSC are located in adult tissue such as the bone marrow, peripheral blood or adipose tissue, as well as in neonatal tissues such as the placenta and the umbilical cord. Traditionally, MSC have primarily been harvested from bone marrow, which is still a common source. However, in addition to being a painful procedure, isolation can cause infections and rather low cell yields. 2 While MSC are well investigated and used for intravascular infusions since decades, for example in patients with leukaemia, recently further sources of MSC gained researcher's interest. 3 Subcutaneous adipose tissue is a favourable source for stem cells, as it is easily accessible and so-called adipose-derived stromal cells (ADSC) can be harvested by minimal invasive surgery in high numbers. 4,5 Like bone-marrow-derived MSC, ADSC are multipotent and described to differentiate into various cell lineages, such as adipogenic, chondrogenic, osteogenic, myogenic, cardiomyogenic, neurogenic and angiogenic cells. [6][7][8][9][10][11][12] ADSC possess immunomodulating properties and secrete various growth factors and cytokines that support tissue regeneration and neovascularization. [13][14][15][16] Therefore, ADSC are promising not only for tissue engineering 5 but also for the treatment of chronic wounds and other inflammatory diseases [17][18][19][20] ( Figure 1A).
Since Zuk et al. published a harvesting protocol for ADSC from adipose tissue in 2001, they have become the focus of scientific interest in the field of stem cell research. 21 In 2016, Naderi et al.
concluded from a range of publications, that the multilineage potential of ADSC makes them an optimal cell type for use in tissue engineering and clinical applications. 5 However, some studies have described variable outcomes of ADSC applications in clinical trials. 22,23 Differences in harvesting and processing techniques have been well investigated but cannot explain the varying effects of ADSC therapies. 24,25 Consequently, several studies focused on the influence of patient-related factors leading to rather controversial results. While some authors could show a correlation between patient-related factors and cell yield, proliferation and differentiation potential, others could not. 25  ADSC functional properties, such as migratory activity, invasiveness and cell-matrix interactions, has not been sufficiently investigated. 26  The patients were surveyed regarding their lifestyle and weight loss history through a questionnaire. An overview of the patient-specific criteria is presented in Table 1.

| Isolation and cultivation of adipose-derived stromal cells
Adipose tissue from post bariatric surgeries was immediately processed under sterile conditions with approximately 40 ml adipose tissue per T75-flask. The subcutaneous adipose tissue was separated from the skin, cut into standardized pieces smaller than 2 × 2 × 2 mm and washed with phosphate buffered saline (PBS) (Sigma-Aldrich).
Tissue pieces were digested with 1 mg/mL collagenase II (Biochrom GmbH) on an orbital shaker at 37°C for approximately 90 min.
Subsequently, the suspension was centrifuged at 400 × g for 10 min, and the supernatant was then aspirated.

| Viability assay
The WST-8 assay was used for measuring the cell viability. ADSC and 600 nm (background) with a microplate reader, and the percentage increase from 24 to 72 h in absorbance was calculated.

| Migration assay
To assess ADSC migration, 5 × 10 4 ADSC were seeded in 100 μl The brightness and the contrast of the images were increased for better visibility.
Then, spheroids of 1,000 ADSC were prepared using the hang-  To ensure the remaining within the linear range of signal amplification the procedure of Wood et al. and to establish compensation values, the manufacturer's instructions were followed. 28 The cut-off value for the positivity was defined by comparison with unstained cells using a histogram, with positive cells being considered to have a signal intensity above the signal intensity of unstained cells. Viability was not measured via flow cytometry, and also the signal-to-noise ratio, the linear range and the coefficient of variation were not determined since these parameters were considered to be dispensable for this study.

| Flow cytometry
Fluorescence minus one controls were included and reproducibility between the experiments was ensured according to the guidelines for the use of flow cytometry and cell sorting by Cossarizza et al. 29

| Statistics
Statistical analyses were performed with GraphPad Prism (GraphPad Software   Figure 1D).

| Functional cell assays
Functional cell characteristics of all ADSC samples were compared through viability, migration, transmigration and sprouting assays.
Significant donor-dependent differences in ADSC proliferation whereas patients 2 and 7 showed low activity in all assays).

| Differentiation assays
Differentiation potential of ADSC from different patients was compared by adipogenic and osteogenic differentiation assays.

| Patient-related factors influencing cell function
To determine the effects of patient-related factors on cell function,

| Influence of BMI on cell function
Since ADSC were isolated from patients who lost a significant amount of weight, it was possible to investigate the influence of the BMI and BMI change on ADSC function. The BMI at the time point of ADSC isolation showed no correlation with ADSC viability, migration, transmigration and sprouting ( Figure 5A). However, the BMI difference was significantly and negatively correlated with the proliferative (p = 0.0305), migratory (p = 0.0153) and transmigratory (p = 0.0119) potential of the ADSC ( Figure 5B).
ADSC from patients with the highest weight loss leading to a high BMI difference showed a lower activity compared with those from patients with a lower BMI difference. The maximum BMI that patients had in the past showed a moderate negative but not significant correlation with the viability, migration and transmigration of ADSC ( Figure 5C). There was no correlation between BMI at time point of ADSC isolation, maximum BMI in patients past nor BMI difference and differentiation potential.
Neither a correlation between CD142 expression and BMI could be shown.

| Telomere length and mtDNA copy number
The relative telomere length and the relative mtDNA copy number and their effects on functional ADSC activity of all donors was determined using quantitative real-time PCR. The relative telomere length and the relative mtDNA copy number were notably different among ADSC, although there was no detectable dependence on the investigated patient-related factors. Considering age (≥45 years n = 9, <45 years n = 10) and sex (female n = 14, male n = 5) as well as the method of weight loss, no significant difference between the groups was observed regarding relative telomere length or relative mtDNA copy number ( Figure 6C). BMI at the time point of isolation, BMI difference due to weight loss and the maximum BMI did not correlate significantly with telomere length or relative mtDNA copy number ( Figure 6D). In addition, telomere length and mtDNA copy number did not correlate with the viability migration, transmigration or sprouting activity of ADSC ( Figure 6A and B).

| Cytokine secretion
The secretion of inflammatory factors and growth factors was analysed using a multiplex ELISA array. Although we noticed differences in the secretion level among different patients for specific factors, there was no patient or group of patients with constant higher secretion. Additionally, we could not determine a correlation between the investigated patient-related factors or the observed cell functions and the secretion levels ( Figure 1E).  teoarthritis, and although all had positive outcomes, the results were highly varied regarding satisfaction and complication rates. 32 Also, heterogenity of functional surface marker expression is described.

| DISCUSS ION
For example, tissue factor (CD142) expression varied among ADSC of different donors, leading to thrombotic events by intravascular application and therefore requires detailed investigation. 33,34 In the present study, we isolated ADSC from 20 different patients during abdominoplasty after massive weight loss and analysed the effects of patient-related factors on ADSC functional properties.
Differences in ADSC functional activities that are likely of great importance when using ADSC in regenerative medicine, such as viability, migratory and sprouting behaviour as well as differentiation potential were compared with reference to patient-related factors. In our analysis, no correlation with sex was observed for viability, transmigration, sprouting activity and differentiation potential of the ADSC, except for migration. Similarly, in the literature, sex is not regarded as an influencing factor on ADSC yield or proliferation. [36][37][38] Since the only correlation between sex and ADSC characteristics was that the migratory capacity of ADSC from female donors was significantly higher that of ADSC from male donors, we conclude that sex only has a minor effect on ADSC function.
In contrast, in many studies, age is described as a potential influencing factor on the proliferative and migratory potential of ADSC. [39][40][41][42][43] For example, ADSC of older patients were shown to have a decreased differentiation potential and a lower cell yield than those from younger individuals. 39 The BMI difference due to weight loss proved to be the strongest influencing factor investigated, with a significant negative correlation observed for the proliferative, migratory and transmigratory potential of ADSC, leading to the assumption that the ADSC of do- ADSC-conditioned medium was found to promote fibroblast proliferation and migration and thereby support wound healing, restore the inflammatory balance and support lymphangiogenesis. [54][55][56] Although we investigated the secretome of the ADSC and noticed patient-depending significant differences for the secretion levels of selected proteins, we could not determine one specific factor explaining the shown functional differences. Additionally, there was no direct correlation between BMI or BMI differences due to weight loss and the analysed cytokines and growth factors. Therefore, we conclude that the interaction between patient-depending factors, inflammation, caloric restriction and ADSC function is based on complex processes and requires further studies with a larger number of patients to be fully understood. Future studies focusing on cellular differences due to the weight loss and their impact on regeneration for instance in an in vivo animal model or an analysis of the RNA profile will help to confirm our conclusion. Mitochondria are important for cellular energy production and cell death, and their number and function are associated with several aging-related diseases. 63  In addition, mtDNA copy number did not correlate with cellular function within our assays.
Interactions among patient factors confounding the effects of ADSC need to be taken into consideration as limiting factors of this study as well as the absence of a patient group with normal weight without distinct weight loss in the past.

| CON CLUS ION
The results of the present study showed significant patient-specific differences in ADSC functions, such as viability, migration, transmigration and sprouting. Within this study, we could show that age, sex or method of weight loss only had a small effect an ADSC functions.
Other influencing factors, such as smoking, diabetes, infections in the surgical area before abdominoplasty, hypertension, hypothyroidism and patient's medication did not correlate with measured cell function. Similarly, BMI at time point of ADSC isolation showed no correlation whereas surprisingly the strongest influencing factor was the BMI difference due to weight loss, with a significant negative correlation observed for the proliferative, migratory and trans- writing -review and editing (equal).

ACK N OWLED G EM ENT
We would like to thank Stefan Fleischer and Julia Christ for their excellent technical support. The present work was performed in fulfilment of the requirements for obtaining the degree 'Dr. med.' for DS. Open Access funding enabled and organized by Projekt DEAL.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.

E TH I C S S TATE M E NT A N D CO N S E NT S TATE M E NT
All procedures performed in studies were in accordance with the