December 12, 2014

Priming of hMSCs to Improve Potency

By Iain Farrance, Priya Baraniak, and Jon Rowley. RoosterBio.

In this blog, we will present internal data and information on priming RoosterBio’s bone marrow derived human MSCs (hMSC) with pro-inflammatory molecules and the impact of these priming protocols on hMSC immunomodulatory function and angiogenic cytokine secretion.

Human Mesenchymal Stem/Stromal Cells, or hMSCs, are key components of future therapeutics, engineered tissues, and medical devices. There are currently over 400 clinical trials investigating hMSCs as therapies (1). The trials have produced some promising results, with hMSCs generally deemed safe, and in some cases effective (2). It is believed that these versatile cells achieve their biologic and therapeutic effects by secreting a plethora of biomolecules (referred to as the MSC secretome) that moderate a variety of processes including angiogenesis, immunosuppression, and overall “tissue repair” (3-6). As the secretome is one of the likely Mechanisms of Actions (MOA) of hMSC therapies, there is a significant amount of recent work on engineering hMSC preparations to enhance secreted factors by genetic modification, by culture strategies, or by engineering the hMSC microenvironment (5, 7-11). In addition, a recent ISCT paper (12) advances the concept of “priming” hMSCs by exposing the cells to pro-inflammatory cytokines prior to implantation.

Thus, priming of hMSCs can have two primary purposes:

  1. To assess human MSC preparations in vitro as recommended by the ISCT and the FDA (12, 13), and
  2. To enhance hMSC potency (survival, immunosuppression, homing) prior to implantation (8, 9, 14).

As part of our standard quality control (QC) testing, RoosterBio analyzes the immunosuppressive capability of our cell lots through priming with IFN-γ. The hMSC response (i.e. immunomodulatory potential) is reported as a measure of indoleamine 2,3-dioxygenase (IDO) activity, determined by measuring the amino acid kynurenine in the culture supernatant.  The IDO enzyme converts L-tryptophan to N-formylkynurenine (or kynurenine), an immunosuppressive molecule that acts as an inhibitor of immune cell proliferation - including T cells (12, 15, 16).  Testing every hMSC lot for inducible IDO activity provides a quality assurance that the cells we release have some level of functional potency as it relates to immunomodulation– which we consider a key quality attribute of hMSCs.

While researchers are beginning to implement testing of hMSC preparations for inducible IDO activity prior to implantation, few are looking at the impact of priming on other hMSC functions.  Here, we present information on priming of RoosterBio’s hMSCs with IFN-γ ± TNF-α across multiple lots and donors and the impact of such treatment on hMSC IDO activity and angiogenic cytokine secretion.  The goal of this blog post is to demonstrate that priming has impacts on several functional properties of hMSCs, and that researchers should consider priming regimens to (a) understand the potency of their specific cell products, especially in inflammatory environments, and (b) to potentially increase potency of these cell products upon therapeutic administration.


Materials & Reagents
Cell culture reagents were purchased from Life Technologies, chemicals and reagents for kynurenine measurement were from Sigma, and cultureware was from Corning.  Other products are: Bone Marrow-derived human MSCs (BM-hMSC, part # MSC-001, RoosterBio) and RoosterBio High Performance Media kit (part # KT-001).

Table 1: Experimental design.

hMSC Priming

November 6, 2014

Current Bottlenecks in MSC Research: MSC Misconceptions - Part II
We blogged recently about Mesenchymal Stem/Stromal Cell (MSC) Misconceptions that are holding the translational cell therapy field back, as identified by Donald Phinney and Luc Sensebé.  Since we have come to market with our own hMSC product lines, we have spoken with hundreds of MSC researchers and engineers, and we have compiled our own set of misconceptions that we think build off of Dr. Phinney’s and Dr. Sensebe’s initial concept.  This blog post is to share some of the market-based feedback that we have received.

To the list that was published in Cytotherapy, we would like to contribute the following list to the conversation:

1. Tracking MSC passage number is an accurate and reliable means of tracking cell age and standardizing experimental workflow
In many research laboratory environments, cellular age is most often tracked by the number of times a cell has been passaged; however, Passage Number is quite imprecise and not very acceptable as one gets into regulated environments such as translational clinical activities.  It is generally accepted that tracking the Population Doubling Level (PDL) or Cumulative Population Doublings (CPD) of primary cells is a best practice on understanding cellular age in vitro Since it is well documented that PDL impacts hMSC function (see here, here and here), in order to drive consistency into experiments, it has become a best practice to perform experiments or develop products with cells in a consistent range of population doublings where the cell function of interest is still robust.  Furthermore, regulatory agencies are beginning to require reporting of PDLs, or at least cell seeding and harvest densities, for primary cells intended for therapeutic use.  In an effort to drive adoption of PDL tracking and reporting, we’ve created a Best Practices Educational Powerpoint, and free, easy-to-use PDL calculator worksheet we’re happy to share with colleagues.  For your copy, just email us at or subscribe to our blog!

2. Performing experiments with one MSC donor and/or lot is adequate for publication and moving forward with pre-clinical studies
Despite indications of clinical effectiveness of MSCs, there is repeated news of the failure of high-profile MSC trials to demonstrate efficacy in a number of therapeutic applications.  It has been suggested that the large amount of intra- and inter-donor variability in the MSC populations used in these trials may be responsible for their falling short of expectations despite highly encouraging in vitro and in vivo pre-clinical data.  Thus, to ensure the robust production of functional MSC products over a range of applications, experiments should be conducted and systems validated with MSCs from several donors It has been reported that best practices to qualify a manufacturing process should include “at least 3-5 donors”, and it is likely that proper Validation will require many more, and that donor selection may be required (i.e. not every donor will work in the manufacturing process). This is why we, at RoosterBio, believe in providing a number of donor MSC lots, ranging in age and sex, for use in our customer’s research and development experiments.

3. MSCs accelerate cancer…..MSCs can combat cancer

October 24, 2014

Current Bottlenecks in MSC Research: Widespread MSC Misconceptions

We blogged several months ago about bottlenecks in the bioprocessing of Mesenchymal Stem Cells that are impeding their clinical translation.  While the development of robust and scalable manufacturing methods, reduced cost of goods, and implementation of solid Quality Systems are all necessary for increased clinical use of MSCs, there are also current misconceptions surrounding MSCs, rooted in decades-old science, that are holding the field back.  I recently came across a paper from last year on this topic and decided to put it forth for discussion here – with a few targeted opinions from us scientists at RoosterBio.  It is my hope that you’ll provide your own opinions on the misconceptions detailed here, as well as your suggestions on other misconceptions you think could be holding the MSC field back.

The authors of the 2013 Cytotherapy Paper: Mesenchymal stromal cells: misconceptions and evolving concepts, Donald Phinney and Luc Sensebé, identify six major misconceptions that have persisted over the years, despite widely-accepted paradigm shifts on MSC nature and function.  Here, I will summarize four of these misconceptions and add our take to them.
Four of the misconceptions identified by Phinney and Sensebé:

1. MSCs isolated from different tissues are equivalent
Initially isolated from bone marrow in the 1950s, MSCs were then discovered in adipose tissue, and have since been found in a number of tissues including, but not limited to: Wharton’s jelly, umbilical cord blood, placenta, amnion, and dental pulp.  While MSCs from all these sources are somewhat similar in surface profile marker expression, phenotype, and gene expression profiles, their functionality, in terms of differentiation potential, immunomodulatory activity, and paracrine factor secretion, can vary widely depending on the tissue of origin.  ** Given that MSCs from a single tissue and donor are not equivalent (see below), it comes as no surprise that MSCs from different tissues vary in function! **

October 2, 2014

Rapid and Economic Generation of hMSC Spheroids for Macroscopic Tissue Biofabrication

Mesenchymal stem cells (MSCs) aggregated into three-dimensional (3D) cellular spheroids are a potent configuration for cell therapy and tissue engineering research and product development (5), and cellular spheroids are a preferred format for many bioprinting applications (10).  Cellular spheroids are essentially micro-tissues that can be manufactured as standardized “living materials” with certain controllable, measurable, and evolving material properties (10). Studies have shown that MSC aggregation into spheroids yield improved in vitro biological functionality over MSCs grown as 2D monolayer; likely due to the 3D tissue-like structure resembling the native configuration of cells in vivo with a microenvironment that allows for direct cell-cell signaling and cell-matrix interactions. MSC spheroids demonstrate enhanced cartilage, bone, and fat differentiation, as well as increased paracrine factor secretion over 2D MSC cultures (1-7). In vivo administration of hMSC spheroids has also showed enhanced therapeutic properties in pre-clinical models of myocardial infarction, bone and cartilage repair, and limb ischemia (3, 5, 8).  

Traditionally, aggregates were formed using suspension culture in spinners or shake flasks or in hanging drop cultures (5). The advancement of technologies has allowed one to quickly and easily generate large numbers of spheroids consistent in size and shape using forced aggregation in micro-wells (AggreWells, Stem Cell Technologies), or using liquid handling automation and 96 or 384 well hanging-droplet plates. There are also tools available that allow researchers to mold micro-tissues into interesting shapes such as rods, toroids, honeycombs, or whatever one can dream up (12-13).  While there are multiple methods for creating hMSC micro-tissues, the biggest challenge is reproducibly growing up sufficient hMSCs to create enough spheroids to start an experiment.  For example, if a researcher needs 10,000 spheroids with an average of 1000 cells per spheroid, then he/she will need at least 10 million cells to begin the experiment, which can take weeks to grow (see process flow diagrams below).  If he/she wishes to use 5000, or 10,000 cells per spheroid, then he/she will need 50 million or 100 million cells for his/her experiment.  This volume of cells has traditionally been very costly and time consuming to generate.

This Application Blog Post will provide a simple protocol to rapidly and economically generate tens of millions of high quality hMSCs so that researchers can minimize their time spent on routine cell culture and maximize their effort on performing hMSC spheroid-based experiments.

September 24, 2014

How Quickly are Cell-based Products Really Developing? Thoughts from the IBC Cell Therapy Bioprocessing Conference

Last week was the 4th Annual IBC Cell Therapy Bioprocessing Conference.  IBC (home of the BioProcess International conference) was the first conference organizer to dedicate a focused meeting on Cell Therapy Manufacturing Technologies 4 years ago.  Since the first conference in 2011, the growth in the field, and the conference, has been amazing.  The attendance has grown from less than 90 in year 1 to over 200 this year.  The content has also evolved heavily over the last 4 years, demonstrating a high level of sophistication and maturity in a field that seems “early stage” to those looking in from the outside.  The talks this year increased in the amount and quality of data presented. Topics included the impact of automation on the simplification, streamlining, and cost reduction of autologous therapies, the use of Quality by Design (QbD) in bioreactor scale-up and analytical development, advances in tissue engineering and biofabrication techniques, and even 2 year data on marketed products.   Phil Vanek, the General Manager of GE Healthcare’s Cell Therapy business, summed it up during his talk where he stated that: GE is interested in 1) big problems, 2) compelling clinical data, and 3) opportunities for “industrialization”, and “Cell Therapy/Regenerative Medicine has all three”.

Various cell manufacturing and processing devices seen throughout the exhibits at IBC's
4th Annual Cell Therapy BioProcessing Conference - No BioPrinters (yet!)
There are many signs that the Cell Therapy field is moving much faster than the protein therapeutics field before it and demonstrating rapid progress.  What we have here is a traditional case of  “standing on the shoulders of giants”, which has been paraphrased on Wikipedia as "discovering truth by building on previous discoveries”.  

September 10, 2014

Scale-up Production of hMSCs: Highlights from the BioProcess Summit Cell Therapy BioProduction Sessions – Post 2 of 2

Fall is almost here, and that means it is Cell Therapy BioProcessing and Manufacturing conference season.  This year it started a few weeks earlier as conference organizer CHI put together a Cell Therapy BioProduction session as part of their Annual BioProcessing Summit in Boston from August 18-22.  In our last post (Scale-up Production – Post 1 of 2), I summarized some of the cool technologies that vendors had on display, as well as some of the poster highlights.  In this post, I want to highlight just a few talks that were focused on manufacturing, scale-up and Cost of Goods of allogeneic cell therapies.  There were several other great talks, but I just wanted to focus on these three due to topic and brevity.

Manufacturing, Cost of Goods, and Unprecedented Stem Cell Process Yields:
On the first day, we had a dynamic duo from Loughborough University give a pair of excellent talks.  Experienced Manufacturing Engineer David Williams gave a great talk on precision manufacturing of living products, highlighting the challenges of working with the inherent variability that comes with primary cell culture.  Dr Williams is the Director of the Center for Innovative Manufacturing in Regenerative Medicine that “works to equip the regenerative medicine industry with manufacturing tools, technologies and platforms by considering the ‘right therapy, right patient, right time’ supply chain from end to end.”  In his talk, he highlighted the need for solid quality characteristics so you know exactly what you are manufacturing – and can do it “again, and again, and again, and again….”.  He points out that, without knowing what characteristics are important for your product (the identity and functional potency of your cells), you: can’t manufacture to specification, can’t scale up, can’t implement new raw materials in your process, can’t transfer manufacturing to another facility, and can’t reduce COGS through process optimization.  Hearing David talk is always a reminder of how important the basics are.

September 5, 2014

Scale-up Production of hMSCs: Highlights from the BioProcess Summit Cell Therapy BioProduction Sessions – Post 1 of 2

Closed system 10 layer and media bags are 
now readilyavailable from several vendors.
Fall is almost here, and that means it is Cell Therapy BioProcessing and Manufacturing conference season.  This year, it started a few weeks earlier as conference organizer CHI put together a Cell Therapy BioProduction session as part of their Annual BioProcessing Summit in Boston from August 18-22.  It is always great to see new conferences including Cell Therapy content, as it shows the maturation of the field.  There is a now a “market” that the organizers believe is worth creating content for.

CHI was kind enough to invite RoosterBio to give the kick-off presentation, so I was able to get re-immersed on all that is new and improved in Cell Therapy Manufacturing Scale-up.  This blog post is meant to share a few of the interesting observations from the meeting and will focus on highlights from the vendor exhibits (i.e. product innovations) and the posters.  The next blog post will share some take-homes from the talks.  There were indeed some valuable and exciting new reports that I want to communicate. 

One key observation is that, while it is still my belief that most of the Allo-products are currently manufactured in 10-layer culture vessels (see above pic), most of the presentations focused on the next generation bioreactor-based processes.  There will continue to be a major shift over the next few years to more automated platforms such as these.

Products on Display for BioProduction of Therapeutic Cells

I always find it worth noting what products the vendors have on display, as they will be developing products based on requests from the market – so more new products displayed means "growing market", which turns into better tools available for everyone.  Interestingly, even at a general BioProcessing conference, there were several booths with Cell Therapy-focused products, and there were 3-5 posters (out of maybe 30) on the scale-up and processing of human MSCs (hMSCs). 

There were a few product innovations and focused product areas among the vendors that I want to highlight here (and we are not getting paid for this, I promise – no sponsorship at all).
Ready-to-Use Microcarriers from Pall

·         Pre-sterilized and ready-to-use microcarriers.  Both Pall/Solohill (see pic) and Corning now offer these products in bottles, but more importantly, in closed system bags ready to seed into a bioreactor system.  Microcarriers in the past had to be prepared and autoclaved by the end user, creating work and yet another set of variables to control when trying to implement this technology.  By providing pre-sterilized microcarriers that are QC’d for efficient cell attachment (I am assuming this QC step; I will try to confirm that), this takes one less process step out of the hands of the process development scientist and makes implementation that much simpler.  This is an important advancement in the field.

·         PBS Biotech reported hMSC densities (expanded in their bioreactors) consistently north of 1 million cells/mL, and up to 3 million cells/mL in a serum containing media.  These numbers are a good 10-fold greater than any number published or presented just 5 years ago, and the highest and most consistent I have seen presented to date.  It is unclear if the reason they achieved these targets was due to the microcarrier and media combination, their novel low shear bioreactor design, or just plain good bioprocess engineering - likely a combination of all three.  In any case, it is a significant achievement, and it demonstrates that it can be done.  Other vendors will now be trying to beat it, and cell therapy companies will be trying to implement it.

·         Vendors are beginning to focus, at least some, on the post-expansion/downstream processing (microcarrier removal, cell concentration) of the cells as well.  Millipore had a presentation and a poster that discussed microcarrier removal using single use filters and the concentration of hMSCs post-harvest using scalable tangential flow filtration (TFF) technology – both with good post-processing viability and recovery.  See poster summary below.  Downstream processing continues to be an under-appreciated aspect of the field and cannot be an afterthought to scale-up culture.  If you scale your expansion to several hundred liters before beginning to think about how you will process the massive cell volumes you have, it could set your program back over a year while these technologies are developed and integrated into the manufacturing process.  It is good to see these aspects of manufacturing get some focus here.

Poster Highlights: 

August 25, 2014

Best Practices in MSC R&D: Addressing Donor Variability within your Experimental System

Human MSCs are the single most used cell source for tissue engineering and regenerative medicine applications, and clinical trials involving hMSCs have outpaced all other cell types in recent years (see here and here).  However, despite indications of clinical effectiveness (see here and here), there is repeated news of the failure of high-profile MSC trials to demonstrate efficacy in a number of therapeutic applications (see here, here, here, here and here).  It has been suggested that the large amount of intra- and inter-donor variability in the MSC populations used in these trials may be responsible for their falling short of expectations despite highly encouraging in vitro and in vivo pre-clinical data.

A team led by Steve Bauer at the US FDA has reported that large variations in proliferation, morphology, differentiation capacity, and cell surface marker expression profiles exist within any population of MSCs and that these intra-population heterogeneities may arise as a result of long-term in vitro culture and the in vivo microenvironment (Free article available here.)  In addition, their work has demonstrated that there are inherent differences in MSCs from donors of similar age, and they have noted the “potential for other donor-related factors in MSC biological variability, which may play a role in their clinical usefulness or performance in various model systems.” Other research groups have also corroborated donor-related differences in MSC function, including in response to stimuli, such as challenge with inflammatory cytokines (see here and here).  A review article on developing cell therapy manufacturing processes reinforces that several donors should be tested prior to implementing; 1) changes in media composition (such as serum reduction/elimination or addition of growth supplements), 2) extensions of the product dose population doubling level (PDL), or 3) changes in lot size during scale-up.

July 7, 2014

Best Practices in MSC Culture: Tracking and Reporting Cellular Age Using Population Doubling Level (PDL) and not Passage Number

Originally Published: July 7, 2014
Updated October 2019 by: Taby Ahsan, Ph.D., Vice President of Analytical, Process & Product Development & Katrina Adlerz, Ph.D., Scientist, Development

What is Population Doubling Level and Why is it Important?
Population doubling level (PDL) is the total number of times the cells in a given population have doubled during in vitroculture. It is well documented in the literature that cell phenotype and function can change the more times cells replicate in vitro. Regulatory agencies have also specified that cellular age should be tracked during manufacturing and that some criteria should be used to set an acceptable upper limit for production. 

Often, cellular age is tracked by the number of times a cell has been passaged. However, passage number is imprecise because different labs may use different initial cell seeding densities which affect the number of times cells divide in culture. It is generally accepted that tracking the population doubling level (PDL) or cumulative population doublings (CPD) of primary cells is best practice for reporting cellular age in vitro

The goal of this blog post is to explain: 1) how passage number and PDL are related 2) how varying cell culture techniques can create a divergence in the reporting of Passage Number compared to PDL and 3) provide guidance and tools to help labs adopt the best practice of tracking PDL of cell cultures to help bring standardization to their own experimental protocols as well as to the field.

Regulatory Guidelines Propose Tracking Population Doubling Levels
There are pharmaceutical regulatory guidelines that address tracking cellular age in vitro. For example, ICH Q5D, Derivation and Characterization of Cell Substrates Used for Production of Biotechnological/Biological Products, states “For diploid cell lines possessing finite in vitro lifespan, accurate estimation of the number of population doublings during all stages of research, development, and manufacturing is important.” 

Another guidance, Points to Consider in the Characterization of Cell Lines Used to Produce Biologicals, points out: “The population doubling level of cells used for production should not exceed an upper limit based on written criteria established by the manufacturer.” This suggests that regulators will ask product developers to define experimentally, with support from data, the maximum PDL that will be acceptable for clinical use.

Population Doubling Level and Cell Function
Regulatory guidelines are only one reason to keep track of PDL. There are multiple papers that specifically discuss cellular age of MSCs and changes in phenotype and function, for example: 
·      Nikbin shows loss of adipogenic and osteogenic differentiation of MSCs with increasing cumulative population doublings here1.
·      Lo Surdo and Bauer show here2that while flow marker expression is stable, there is a decrease in proliferation rate and a loss of adipose differentiation in hMSCs from passages 3 to 7.
·      Le Blanc retrospectively proposes here3that MSCs from passages 1 or 2 are more therapeutically functional in GvHD than MSCs from “later” passage 3 or 4 cells (the later passage cells were also cryopreserved). 
·      Braid shows here4that proteomic drift can occur at higher population doublings.

Since it is well-documented that PDL impacts cell function, in order to drive consistency in experiments, it has become best practice to perform experiments with cells in a similar range of population doublings where the cell function of interest is still robust– whether that function is secreted cytokines, multi-lineage differentiation, or the ability to modulate immune function. For bone marrow-derived MSCs, most researchers report performing experiments with cells in the passage range of 4 to 6. With a traditional MSC culture protocol where there are 2.5 - 3 population doublings per passage, this results in MSCs in a PDL range of 12 - 18. For umbilical cord-derived MSCs, typically there are 5 - 5.5 population doublings per passage, such that many experiments are with cells in the PDL range of 25 - 30.

Why Passage Number Is Not Enough
The process of culturing cells, including MSCs, can vary greatly between labs and dramatically impact the number of population doublings per passage. To illustrate this, we will look at 3 representative culture processes listed below (and outlined in the table below):

1. A “traditional” MSC culture method of seeding cells at a density of ~5,000 cells/cm2and harvesting at ~80% confluence (which is usually ~20,000 cells/cm2) will lead to MSCs doubling twice per passage (5,000 to 10,000, then 10,000 to 20,000 – or 2 doublings per passage);
2. A lower seeding density of 1,250 cells/cm2will produce 4 doublings per passage (assuming the same harvest density);
3. And a hyper-low seeding density of 78 cells/cm2will produce 8 population doublings per passage.

Since the seeding density and harvest density can vary greatly between labs, reporting passage number is not a standardized means of reporting cellular age. For example, if the goal is to use cells at a population doubling level between 12 and 18, for well controlled experimental and manufacturing processes, every time an experiment is performed, cells should be within 6 - 8 passages from culture process #1 above, within 3 - 4 passages using the intermediate-density culture process #2 above, and only at passage 2 using the hyper-low density seeding of cell culture process #3 above. We have attempted to outline the impact on cumulative population doublings per passage based on these 3 different methods in the chart below.

Cumulative Population Doubling Level at Varying Seeding Densities with Harvest Density of 20,000 cells/cm2

Passage Number
5000 cells/cm2
1250 cells/cm2
78 cells/cm2

Passage 2 cells from a lab using cell culture process #1 are clearly not the same cellular age as Passage 2 cells from lab culture processes #2 or #3. Furthermore, cell culture is performed on the human’s schedule, not the cells. Cells are often harvested earlier or later due to scheduling conflicts, illness, weekends, etc. These “small” changes in timeframe can lead to large variations in PDL and resulting experimental outcomes. And importantly, if PDL is not tracked, these details are lost, and experimental outcomes cannot be evaluated based on differences in PDL.
So How Do You Calculate PDL?
The ATCC website contains the following: “…Passage number simply refers to the number of times the cells in the culture have been subcultured, often without consideration of the inoculation densities or recoveries involved. The population doubling level (PDL) refers to the total number of times the cells in the population have doubled since their primary isolation in vitro.” MSCs are a rare population in bone marrow and it is difficult to estimate the starting number of MSCs in the initial culture. So, by convention, most labs start counting MSC cumulative population doublings after the P0 cell harvest. Furthermore, PDL is not designed to take into account the number of times these cells have divided in vivo, that is where donor age and health comes into play as another important variable to monitor.

To calculate the PDL of your cell cultures, you can use the equation below:

PDL= initial population doubling level
C= initial cell number seeded into vessel
Cf= final cell yield, or the number of cells at the end of the growth period

Take Home Message
The best way to report cellular age is using PDLs. Well controlled experimental and manufacturing process will use cells within a consistent PDL range. Therefore, RoosterBio reports the exact PDL of each lot of MSCs so that our customers can keep track of cumulative PDL during their own experiments and manufacturing processes. We have also created an Excel Template PDL Tracker that anyone can request for their own purposes (Please email us at 

1.     Bonab, Mandana Mohyeddin, et al. "Aging of mesenchymal stem cell in vitro." BMC cell biology 7.1 (2006): 14.
2.     Lo Surdo, Jessica, and Steven R. Bauer. "Quantitative approaches to detect donor and passage differences in adipogenic potential and clonogenicity in human bone marrow-derived mesenchymal stem cells." Tissue engineering part C: methods 18.11 (2012): 877-889.
3.     Moll, Guido, et al. "Do cryopreserved mesenchymal stromal cells display impaired immunomodulatory and therapeutic properties?." Stem cells 32.9 (2014): 2430-2442.

4.     Wiese, Danielle M., et al. "Accumulating Transcriptome Drift Precedes Cell Aging in Human Umbilical Cord‐Derived Mesenchymal Stromal Cells Serially Cultured to Replicative Senescence." Stem cells translational medicine (2019).