• RoosterBio’s webinar revealed how tissue source, donor variability, and culture format (2D vs. 3D) significantly influence extracellular vesicle (EV/exosome) yield and prospective quality attributes.
• Robust analytics were emphasized, including particle tracking, protein/RNA profiling, and CD73 activity, all supporting regulatory-ready CMC development.
• 3D bioreactors consistently produced more EVs per cell, with bone marrow MSCs showing the highest output and notable differences in RNA cargo and functional markers.

Who?

Whan that Aprille with his shoures soote,
The droghte of March hath perced to the roote,
And bathed every veyne in swich licóur
Of which vertú engendred is the flour

– The Canterbury Tales, Geoffrey Chaucer

Chaucer opened his famous 14th-century epic with an ode to April’s saturating rainfall that begets the majesty of flowers. RoosterBio kicked off April 2025 with a webinar to report on the floodgate of new data pouring in from its analytics team. To obtain flourishing mesenchymal stem/stromal (MSC) cell cultures with abundant extracellular vesicles (EVs; sometimes referred to as exosomes), the company continues to optimize toward the most efficient process parameters and ingredients. Yet it also naturally follows that RoosterBio has grasped the imperative to properly characterize such products with analytical assays that are standardized and transferable.

As extracellular vesicles become an exciting new platform for diverse unmet clinical needs, means to qualify their identity, potency, purity, and safety must be assured. The drug substance of MSC-EVs and their critical quality attributes (CQAs) need interrogation with the right analytical tools. Accordingly, we and others are often asked how measurable production process parameters and tissue source choices impact the quality of MSC-derived EVs. Being scientists, we don’t like keeping secrets for very long. That’s the purpose of our recent webinar, and why a large number of viewers recently tuned in. We eagerly invite you to do the same by watching the webinar here.  …Unlike Chaucer & English Lit 101, no annotated translation from Middle English is necessary! For an abbreviated summary of this webinar, read on.

Drs. Elie Zakhem, PhD, (Associate Director of Analytical Development and Services) and Stephen Lenzini, PhD, (Senior Scientist) were well-equipped to tackle this question for our audience. Elie oversees the efforts to develop and qualify analytical methods for MSC and EV characterization. Prior to this role, he was the manager of Process development at RoosterBio, leading our work to develop scalable manufacturing processes for MSCs and EVs’ bioproduction and purification. Before joining RoosterBio, Stephen earned his PhD in Bioengineering at the University of Illinois at Chicago, where he published extensively on EV biology and bioprocessing.

Why?

Founded in 2013, RoosterBio’s original mission was to radically simplify MSC production and offer scalable, GMP-ready systems for cell expansion. This original mission nicely dovetailed with the subsequent explosion in scientific interest to clinically translate EVs/exosomes. Dr. Elie Zakhem reported that the company was thus well-poised to pivot toward an upgrade of its in-house expertise and novel solutions to contend with the biomanufacturing complexities found with this new modality.

Extracellular vesicles/exosomes are prospective advanced therapeutics derived from the secretomes of MSCs and other cells or biofluids. MSCs comprise the active ingredient in at least 1500 posted clinical trials worldwide and were recently approved as a therapy for juvenile GvHD by the FDA. [1] The conditioned media from MSCs that include EVs may embody many of these cells’ putative properties, valuable to regenerative medicine. [2, 3, 4] MSC-EVs are thus increasingly validated in vivo, showing promise in a variety of preclinical studies. [5] Further, this modality is potentially more easily storable, injectable through more diverse routes of administration, freer from risks related to suboptimal cell viability and infection, and more able to distribute widely (e.g., without embolism formation) on account of their smaller size. [6, 7, 8, 9]

Extracellular vesicle surfaces are engineerable for tissue targeting and/or amplified signal transduction with multivalent ligands, and their interiors can be packaged with pharmaceutical cargoes, enabling more precise control over MOA and/or in vivo targeting/retention. [10, 11, 12, 13] The natural biogenesis of EVs lends them possible advantages beyond artificial LNPs as a drug delivery system for both large and small molecules, perhaps being less prone to wayward immune reactions. [14][15] Although EVs in well-conducted human trials have yielded very few, if any, serious adverse events, [16] liposomes and/or LNPs are not without their own risks, [17, 18] and their scalability and manufacturability is not necessarily straightforward. [19] Surprisingly, Elie noted that the number of PubMed publications related to EVs/exosomes began growing geometrically since ~2011, and now exceeds the annual tally of papers querying to mAbs or liposomes. The count of newly posted human trials with EVs also consistently exceeds dozens per year as of this writing.

Dr. Zakhem then brought CMC (Chemistry, Manufacturing and Controls) to the audience attention. CMC information is required by government agencies like the FDA for new investigational drug filings (the “IND”). It tells the story of how the product is to be manufactured in a standardized and reliable manner. Unlike traditional drugs, EV therapies face unique bioproduction challenges that directly shape the CMC section of IND regulatory filings.

An IND CMC Section evidently cannot write itself, but RoosterBio has now made this task much less onerous with modular, customizable, bioreactor-compatible workflows that accelerate the transition from research to clinical-grade production. These are undergirded with high-performance GMP-grade working cell banks and bioprocess media that are tied to Type II US FDA Biologics Master Files (MF) for easy cross-referencing. Specific to EVs, Rooster also launched drop-in products like a low-particulate EV collection medium (RoosterCollect™-EV) and AgentV™-DSP, which may dramatically increase final yields and purity of MSC-specific EVs, where vialed EV doses may even approach mAb-like COGS.  [20]

“Manage what you can measure…Expect what you can inspect.” An EV product developer might keep these pithy adages in mind through the assignment of right-fitting CQAs that align with the substance’s Target Product Profile (TPP). But what exactly is worthy of measuring and inspecting to define as a CQA? There are only so many terabytes of analytics data (and person-hours) that can stream from continuous process monitoring. The answer to this dilemma is to define the CQAs by pinning down the controllable parameters that can directly ramify into product identity, potency or safety during a manufacturing process. [21, 22, 23, 24]

There are presently(!) no FDA-approved EV/exosome products to offer a model of unequivocal success. Nevertheless, Dr. Zakhem explained that MISEV (Minimal Information for Studies of Extracellular Vesicles) can fortunately provide a solid foundation to facilitate the first steps on CQA definition. To be clear, MISEV is *NOT* a checklist for regulatory guidelines alignment, nor is it one-size-fits-all rulebook, and certainly not a barrier to innovation. It is best conceived as a set of fundamental questions to advance rigorous best practices in EV science. [25] As such, principles outlined in MISEV can shine light on robust assays and instrumentation to steer realistic progress.

Grounded in empirical science but squarely aimed at viable future products, RoosterBio offers its customers a wide range of cellular and EV analytical characterization services that can facilitate its customers’ analytical method development and standard assay qualification. These span particle size distribution (NTA), survey of surface markers (CD9, CD63, CD81) and prototypical intraluminal cargo proteins (Alix, TSG101), RNA content analysis, residuals of media components (albumin and external nucleic acids), CD73 enzyme activity, cellular potency, to name a few. Dr. Zakhem concluded his section of the webinar to outline goals of RoosterBio’s recent studies.

Because standard quality metrics are essential to successful clinical translation of new therapeutic modalities like EVs, it was necessary to methodically determine whether EV quality is affected by different MSC producer cell types (donor or source) and different 2D or 3D production platforms.

How?

The presentation was turned over to Dr. Stephen Lenzini, who then emphasized the analytical rigor needed to establish CQAs for EVs. The key currency of successful EV upstream (USP) and downstream process development (DSP) is relevant analytics data that both emerges from the process and also shapes it. USP, DSP, and analytics are closely interdependent. Simply stated, robust USP development seeks to answer how to generate EVs, what cell expansion platform to use, and what the process parameters are. The DSP development answers which process parameters are crucial to purification and how to achieve it, and what the key unit operations are. The analytical assays answer how to define the EV product and whether the EVs continue to meet this definition in monitorable quality attributes; reliable assays are the eyes, ears, and nose of bioprocessing.

RoosterBio’s customers and the EV community have naturally been curious about how important process variables affect their prospective CQAs. Thus, RoosterBio set out to thoroughly explore this question experimentally, focusing on the upstream process parameters. For the proposed experiments, RoosterBio’s process development team architected fixed process parameters; a short list of examples of these included of seed train duration (4 days), Growth Medium (RoosterNourish™), Growth Phase Duration (5 days), and Collection Medium (RoosterCollect™-EV).

The fixed parameters aligned with a common workflow template as shown in Figure 1 (below).

Figure 1. Generic upstream process (USP) workflow anchored to all testing of MSCs in these experiments, including the seed train, MSC expansion phase, and last, EV collection phase.

In turn, RoosterBio tested important variable process parameters (Table 1, below):

Variable Process Parameters
Scalable Platform	Stirred Tank, Vertical Wheel, Spinner Flask, 2D Flask
MSC Tissue Source	Bone Marrow, Umbilical Cord, Adipose Tissue
MSC Donor	Donor 1, Donor 2, Donor 3
Collection Day	Day 2, Day 5

Table 1. RoosterBio investigated controllable process parameters that might affect EV CQAs, such as cell expansion platform, MSC tissue and/or donor, and different collection times.

Because MSC tissue subtypes (bone marrow, umbilical cord, adipose tissue), human donor phenotype/genotype, and collection time might affect growth kinetics and measurable attributes of EVs, RoosterBio selected these as key variables to scrutinize. Duration of collection time might also change the features of EVs in conditioned media, so collection times of Day 2 and Day 5 were chosen for comparison. Finally, it’s possible that different types of shear stress or growth surfaces could modulate the number and morphology of EVs via USP, so different scalable cell culture platforms were surveyed from Sartorius, PBS Biotech, and Corning, shown in Table 2 (below).

	Stirred Tank	Vertical Wheel	Spinner Flask	2D Flask
Tested System	Ambr®250 (Sartorius)	PBS Mini (PBS Biotech)	Corning	Corning CellBIND® CellSTACK®
Tested Scale	250mL	100mL	100mL	2-layer (1272cm²)
Maximum Scale	Up to 2000L	Up to 80L	Up to 3L	10-layer (6360cm²)

Table 3. To approach reliable markers of Identity, Purity, and Biofunction/Potency, various instrument and assay platforms were used that yielded relevant readouts to quantify number & size of EVs, report on the presence of canonical EV associated protein & RNA markers, assess sample purity from culture residuals (protein and nucleic acid), and determine if the sample retained biologically relevant enzyme function after DSP (CD73 activity).

How were these variables (biological, timing, mechanical) surveyed with practical analytical assays, and could they perform robustly enough to establish durable CQAs? That is, could these platforms stand as reliable under different conditions without hindering their transferability into different “sets of hands?”  Stephen explained the different methods and instruments to address Identity, Purity and Biofunction, summarized in Table 3 (below).

	Identity			Purity		Function
	Nanoparticle Tracking Analysis (NTA)	Protein Marker Expression	RNA Target Expression	Total Protein Content	Total RNA Content	CD73 Activity
Instrument / Assay	ZetaView®	Capillary Western Blot (ProteinSimple Jess)	Quantitative PCR	Bradford Assay	Agilent Bioanalyzer	AMP Conversion Assay
Readout(s)	(1) Particles per mL (2) Particle Size (nm)	Presence / Absence of CD63, CD9, CD81, ALIX, TSG101	Presence / Absence, Relative Quantity of select RNA targets	Protein Concentration (µg/mL)	RNA Concentration (ng/mL)	Enzymatic Activity per particle (pmol/min per 1e9 particles)
Relevance	Quantity and description of EV material in a sample	EV markers demonstrate presence of EVs in a sample	RNA targets relevant to therapeutic applications can be detected	Provides information about total biological content in a sample, useful for description and evaluation of EV purification state		Measures function of a relevant protein detected in MSC-EVs

Table 3. To approach reliable markers of Identity, Purity, and Biofunction/Potency, various instrument and assay platforms were used that yielded relevant readouts to quantify number & size of EVs, report on the presence of canonical EV associated protein & RNA markers, assess sample purity from culture residuals (protein and nucleic acid), and determine if the sample retained biologically relevant enzyme function after DSP (CD73 activity).

As expected, the data output from this ambitious comparative study was prolific. To find out more, read on, or view the webinar.

What?

It would betray the purpose of a blog for quick consumption to exhaustively show each of the 12+ data panels in the webinar. Nevertheless, here are some highlights.

Dr. Lenzini continued the webinar with a walk-through of particle analyses via the ZetaView NTA. Consistent with our own previous results [26] and others’, [27] 3D culture systems consistently outperformed 2D (i.e., planar culture on Corning CellSTACK® flasks) in particle production, perhaps due to the stimulatory effect of mild shear stress on innate cellular EV bioproduction and release (Figure 2, below).  While particle size remained consistent across culture platforms, it varied by tissue type, with umbilical cord EVs yielding the largest, followed by bone marrow and adipose tissue. The most prolific cell type, measured in particles per cell, were the MSCs form bone marrow. The longer the time is allowed for collection, the more EVs can be harvested. In short, with quantifiable particle concentrations and sizes achieved that are reproducible within their individual production schemes, these results demonstrate that NTA can be an effective part of the toolkit for evaluating future CQAs.

Figure 2. NTA data show there is little difference between 3D platforms (stirred tank, vertical wheel, spinner flask) re: EV productivity, but 3D consistently outperforms cells cultured in 2D flasks. Also, increased collection time (2 days vs. 5 days) yields increased particle counts. While not obvious here, MSCs from bone marrow are more productive when measured in EVs per cell (data not shown), perhaps by as much as 2-3 fold.

Stephen then moved the focus of the webinar to protein and RNA characterization of the EVs prepped from stirred tank 3D culture via bone marrow, umbilical cord, and adipose MSCs. It was found that total protein content of the samples increased between Days 2 and 5, very tightly correlating with number of EVs released under each condition. It is thus plausible that a range of appropriate particles/mL of sample protein might map to a purity-related CQA for prospective EV bioprocesses. To validate the identity of the samples as EVs, canonical exosome markers (CD9, CD63, CD81, TSG101) were detected by capillary Western blot images, as was a putative marker of MSCs and MSC-EVs (CD73). With appropriate controls, this method could thus be a viable method to show that a sample is enriched in MSC-EVs.

The role of RNA as an EV-associated or encapsulated “cargo” with potential signaling significance continues to be avidly explored and even debated. However, like protein, there may be tightly and loosely selective species of RNA to be found in preparations of EVs, and relative abundances of these could reflect a metric of purity or potency, monitorable by routine qPCR assays.  RoosterBio found that total RNA content (ng/mL) and particles per ng of RNA varies between EVs from bone marrow, umbilical cord, and adipose MSCs. Not only does total RNA vary, but also individual species of miRNA “cargo” that may be relevant to a therapeutic mechanism. [28] RoosterBio accordingly measured levels of miR-125b-5p (~anti-inflammation), miR-23a-3p (~angiogenesis), miR-221-3p (~cardiac repair), and miR-92a-3p (~neuroprotection) relative to a normalization control of miR-16-5p between MSC-EVs from bone marrow, umbilical cord, or adipose. Abundance of these varied between EVs from MSCs sourced from different tissues. Therefore, since RNA content can vary across different process parameters, it is extremely important to evaluate RNA content that is suitable for the intended therapeutic goals.

Reliable and simple metrics that can report from an EV sample’s potency are highly sought. Thus, Dr. Lenzini next communicated RoosterBio’s recent work involving CD73, which is both a putative marker of MSCs [29] as well as an EV-associated mediator of immune modulation via its Ecto-5’-nucleotidase activity and adenosine generation. [30, 31] CD73 may leverage effects in vivo when in proximity with CD39 activity through adenosine second messengers, or act as a dampener of the extracellular ATP “damage” signal. CD73 enzyme activity was detected in all samples, yet its activity per particle varied according MSC tissue of origin as well as different donors.

This puzzling CD73 activity variability among samples compelled RoosterBio to successfully push the envelope of its NanoFCM instrument to enumerate individual CD73 molecule counts per single EV particle in a novel assay. Stephen showed that his colleagues’ method of calculation yielded stable numbers in each sample across collection times, where number of CD73 molecules expectedly correlated with CD73 enzyme activity. (Figure 3, below).

Figure 3. A new assay counts individual CD73 molecules on single EV particles across samples derived from MSCs of different donors and tissue origins. There is predictable correlation between number of CD73 molecules and CD73 enzyme activity per donor/sample. This method can be transferable to other unique molecular markers on EVs, such as natural or artificial ligands that might govern EV potency. This work supports an award-winning abstract (#37)  highlighted as a podium presentation, “Single Extracellular Vesicle Profiling: Harnessing NanoFlow Cytometry for Quantitative Surface Marker Analysis” at ISCT-2025 by Madeline Cramer, PhD!

Lenzini concluded the webinar by stressing that donor-based variability underscores the need for rigorous screening when selecting MSC sources for consistent therapeutic function. Given the importance of building a CMC-aligned analytical framework for EV products, RoosterBio’s tools and methods are paving the way for regulatory-grade standardization.

Q&A

An energetic question-and-answer session followed Elie and Stephen’s presentations, which included questions such as:

• Is the single particle analysis method specific to the CD73 marker used?
• For particle-per-cell productivity, do you use cumulative or end-production cell density, and do you use viable or total cell density?
• Can RoosterBio qualify assays for GMP testing?
• How do you discern between particles from the media and particles produced by cells, especially given the limitations of bioreactor wash steps?
• What differentiates RoosterCollect-EV media from growth media in terms of reducing background?
• Also, how does confluency compare between Day 2 and Day 5 collections and was the same collection media used?
• Is there a universal EV product, or should the process be tailored to end goals?

Would you like to hear RoosterBio’s answers to these? Please access the webinar recording and listen to find out!

April, Back Full Circle

We all sense something primal in those cool April showers and bright sunshine, first stirring the roots and shoots of new life, which then erupt exuberantly in greenness everywhere, all at once. They have motivated RoosterBio to studiously prepare for its round of spring conferences with exciting news prefaced partly by this webinar. Chaucer wrote that April’s verdant vibe of vitality was a cause to motivate his 30-or-so ragtag pilgrims to set out on the road to Canterbury, as well.

Was Chaucer’s cross-section of medieval society not a bit like us in the EV community? If anything, the writer crafted it as an ode to an “interdisciplinary” bunch of women and men, rich and poor, some with possibly different agendas or baggage. Similarly, our 21^st Century interdisciplinary bioscience gathers unique minds and voices who might not instantly understand one another, and sometimes snicker or debate… all in good fun, of course. But aren’t we all moving in the same direction? And doesn’t this muddle provide fertile ground for inevitable breakthroughs?

For its part on the journey, RoosterBio looks forward to sharing more of its own epic science story as it unfolds.

 

References

1. Giebel, B., A milestone for the therapeutic EV field: FDA approves Ryoncil, an allogeneic bone marrow-derived mesenchymal stromal cell therapy. Extracell Vesicles Circ Nucl Acids, 2025. 6(1): p. 183-190. 10.20517/evcna.2025.02
2. Kordelas, L., et al., MSC-derived exosomes: a novel tool to treat therapy-refractory graft-versus-host disease. Leukemia, 2014. 28(4): p. 970-3. 10.1038/leu.2014.41
3. Timmers, L., et al., Human mesenchymal stem cell-conditioned medium improves cardiac function following myocardial infarction. Stem Cell Res, 2011. 6(3): p. 206-14. 10.1016/j.scr.2011.01.001
4. Giebel, B. and S. K. Lim, Overcoming challenges in MSC-sEV therapeutics: insights and advances after a decade of research. Cytotherapy, 2025. 10.1016/j.jcyt.2025.03.505
5. Elahi, F. M., et al., Preclinical translation of exosomes derived from mesenchymal stem/stromal cells. Stem Cells, 2020. 38(1): p. 15-21. 10.1002/stem.3061
6. Sharma, D. and F. Zhao, Updates on clinical trials evaluating the regenerative potential of allogenic mesenchymal stem cells in COVID-19. NPJ Regen Med, 2021. 6(1): p. 37. 10.1038/s41536-021-00147-x
7. Sun, Y., et al., Mesenchymal stem cells-derived exosomes for drug delivery. Stem Cell Res Ther, 2021. 12(1): p. 561. 10.1186/s13287-021-02629-7
8. Kalluri, R. and V. S. LeBleu, The biology, function, and biomedical applications of exosomes. Science, 2020. 367(6478). 10.1126/science.aau6977
9. Lenzini, Stephen. Big Effects in Small Packages: What Are Extracellular Vesicles, Exosomes, & Microvesicles & Why Are They En Route to the Clinic? RoosterBio Blog 2021; Available from: https://www.roosterbio.com/blog/big-effects-in-small-packages-what-are-extracellular-vesicles-exosomes-microvesicles-why-are-they-en-route-to-the-clinic/.
10. Alvarez-Erviti, L., et al., Delivery of siRNA to the mouse brain by systemic injection of targeted exosomes. Nat Biotechnol, 2011. 29(4): p. 341-5. 10.1038/nbt.1807
11. Kojima, R., et al., Designer exosomes produced by implanted cells intracerebrally deliver therapeutic cargo for Parkinson’s disease treatment. Nat Commun, 2018. 9(1): p. 1305. 10.1038/s41467-018-03733-8
12. Kim, S., et al., Dual-mode action of scalable, high-quality engineered stem cell-derived SIRPalpha-extracellular vesicles for treating acute liver failure. Nat Commun, 2025. 16(1): p. 1903. 10.1038/s41467-025-57133-w
13. Kamerkar, S., et al., Exosomes facilitate therapeutic targeting of oncogenic KRAS in pancreatic cancer. Nature, 2017. 546(7659): p. 498-503. 10.1038/nature22341
14. Omata, D., et al., Effect of Anti-PEG Antibody on Immune Response of mRNA-Loaded Lipid Nanoparticles. Mol Pharm, 2024. 21(11): p. 5672-5680. 10.1021/acs.molpharmaceut.4c00628
15. Kozma, G. T., et al., Role of anti-polyethylene glycol (PEG) antibodies in the allergic reactions to PEG-containing Covid-19 vaccines: Evidence for immunogenicity of PEG. Vaccine, 2023. 41(31): p. 4561-4570. 10.1016/j.vaccine.2023.06.009
16. Van Delen, M., et al., A systematic review and meta-analysis of clinical trials assessing safety and efficacy of human extracellular vesicle-based therapy. J Extracell Vesicles, 2024. 13(7): p. e12458. 10.1002/jev2.12458
17. Inglut, C. T., et al., Immunological and Toxicological Considerations for the Design of Liposomes. Nanomaterials (Basel), 2020. 10(2). 10.3390/nano10020190
18. Wang, J., et al., Recent Advances in Lipid Nanoparticles and Their Safety Concerns for mRNA Delivery. Vaccines (Basel), 2024. 12(10). 10.3390/vaccines12101148
19. Eugster, Remo and Paola Luciani, Liposomes: Bridging the gap from lab to pharmaceuticals. Current Opinion in Colloid & Interface Science, 2025. 75: p. 101875. https://doi.org/10.1016/j.cocis.2024.101875
20. RoosterBio. To 10X & Beyond, AgentV™-DSP Arrives Ready to Rock. RoosterBio Blog 2025; Available from: https://www.roosterbio.com/blog/to-10x-beyond-agentv-dsp-arrives-ready-to-rock/.
21. Carson, J. “Known Unknowns” Affect Critical Process Parameters & Critical Quality Attributes in the Bioproduction Cycle. RoosterBio Blog 2021; Available from: https://www.roosterbio.com/blog/known-unknowns-affect-critical-process-parameters-critical-quality-attributes-in-the-bioproduction-cycle/.
22. Adlerz, K. Critical Quality Attributes (CQAs): Know Their Importance & Limitations in Product & Process Development. RoosterBio 2021; Available from: https://www.roosterbio.com/blog/critical-quality-attributes-cqas-know-their-importance-limitations-in-product-process-development/.
23. Lembong, J. Identify & Define Your Cell Therapy’s Biomanufacturing Approach for Critical Process Parameters (CPPs). RoosterBio Blog 2021; Available from: https://www.roosterbio.com/blog/identify-define-your-cell-therapys-biomanufacturing-approach-for-critical-process-parameters-cpps/.
24. Kirian, R. Maintaining CQAs as Manufacturing Processes are Scaled from 2D to 3D Bioreactor Culture. RoosterBio Blog 2022; Available from: https://www.roosterbio.com/blog/maintaining-cqas-as-manufacturing-processes-are-scaled-from-2d-to-3d-bioreactor-culture/.
25. RoosterBio. The Spirit of MISEV. RoosterBio Blog 2025; Available from: https://www.roosterbio.com/blog/the-spirit-of-misev/.
26. Adlerz, K., et al., Comparison of msc-evs manufatured in 2D versus scalable 3D bioreactor systems. Cytotherapy, 2019. 21(5): p. S58. 10.1016/j.jcyt.2019.03.434
27. Jeske, R., et al., Upscaling human mesenchymal stromal cell production in a novel vertical-wheel bioreactor enhances extracellular vesicle secretion and cargo profile. Bioact Mater, 2023. 25: p. 732-747. 10.1016/j.bioactmat.2022.07.004
28. Arminan, A., Carson, J., Lenzini, S. miRNA, miRNA in the QIAzol®, Where’s the Fingerprint Among Them All? RoosterBio Blog 2023; Available from: https://www.roosterbio.com/blog/mirna-mirna-in-the-qiazol-wheres-the-fingerprint-among-them-all/.
29. Dominici, M., et al., Minimal criteria for defining multipotent mesenchymal stromal cells. The International Society for Cellular Therapy position statement. Cytotherapy, 2006. 8(4): p. 315-7. 10.1080/14653240600855905
30. Hettich, B. F., et al., Exosomes for Wound Healing: Purification Optimization and Identification of Bioactive Components. Adv Sci (Weinh), 2020. 7(23): p. 2002596. 10.1002/advs.202002596
31. Cramer, M. CD73: A Team Player Caught in the “AKT” of Wound Healing & Cell Survival via MSC Exosomes? RoosterBio Blog 2024; Available from: https://www.roosterbio.com/blog/cd73-a-team-player-caught-in-the-akt-of-wound-healing-cell-survival-via-msc-exosomes/.