Building a digital library on research into mineralizing vesicles: a systematic review-based approach

Abstract

This systematic review consolidates current research on mineralizing extracellular vesicles, or matrix vesicles (MVs), including their isolation, characterization, and role in physiological and pathological calcification. We searched PubMed/Medline, Scopus, and Web of Knowledge by employing the keywords “matrix vesicles” or “collagenase-released matrix vesicles” or “mineralizing vesicles” and publishing years from 2000 to 2023. Seventy-one studies met the inclusion criteria. The studies described different experimental protocols, especially with respect to methods for isolating MVs, wherein digestion with collagenase combined with centrifugation was the most used. The studies employed characterization techniques, including the determination of alkaline phosphatase (ALP) and transmission electron microscopy (TEM), to assess the functionality, size, and morphology of MVs. MVs contain key proteins such as ALP, annexins, and osteocalcin, along with calcium and phosphate ions, which are all critical for precipitating apatite. In the studies, evaluation of ALP activity revealed that MVs are more effective for mineralization than their parent cells and, hence, a valuable tool to regenerate bone and to engineer tissues. On the other hand, MVs play an essential role in pathologies, and the studies showed how they contribute to vascular calcification. Despite the therapeutic potential of MVs, isolation methods and characterization protocols vary across the studies, so standardized methods are needed. We have consolidated the data resulting from this systematic review in an open digital library on MVs with free access to all researchers. The users of the digital library can apply filters and taxonomy to find and interconnect the data resulting from the review.

Introduction

Extracellular vesicles (EVs) originate from cells and have a lipid bilayer structure, but they cannot self-replicate because they do not contain a functional nucleus (Welsh et al. 2024; Lötvall et al. 2014). EVs mediate several biological processes and underlie cell-to-cell communication. Researching the structure and molecular composition of EVs originating from different cells opens important avenues for discoveries in basic biophysics and applied fields. For example, biomarkers can be identified, and therapies can be developed. However, EVs are complex, which makes their isolation, characterization, and functional analysis challenging (Welsh et al. 2024; Vaiaki and Falasca 2024).

Matrix vesicles (MVs), a subtype of EVs known as mineralizing vesicles, are specialized extracellular matrix (ECM)-derived vesicles with diameters typically ranging from 50 to 200 nm. MVs originate from the budding of plasma membranes of mineralization-competent cells such as osteoblasts, odontoblasts, and chondroblasts (Ansari et al. 2021; Buchet et al. 2013; Anto et al. 2024) and release lipid- and protein-rich particles into the extracellular milieu (Welsh et al. 2024; Man et al. 2023). MVs enriched with phosphatases, phosphate and calcium transporters, and annexins can accumulate calcium and phosphate ions, causing apatite minerals, the primary component of the bone mineral matrix, to precipitate (Yang et al. 2024). Thus, MVs play a crucial role in bone mineralization, contributing to bone growth and maintenance of structural integrity (Anto et al. 2024; Tamura et al. 2023; Yang et al. 2024). The part played by MVs in cell-to-cell communication has been inferred from the presence of desoxyribonucleic acid (DNA) and ribonucleic acid (RNA), including microRNAs (miRNA), in their composition, which is critical for cellular signaling and regulation (Welsh et al. 2024; Man et al. 2023).

Emerging research has highlighted that MVs are key regulators of bone mineralization and remodeling (Welsh et al. 2024; Zhang et al. 2024; Zhuang et al. 2024), vascular calcification, and cardiovascular health (Ansari et al. 2021; Zazzeroni et al. 2018), suggesting that MVs are involved in pathologies. MVs influence the expression of mineralization-related genes, mediate cellular interactions, and impact ECM remodeling (Li et al. 2022). These properties underscore that MVs are potential therapeutic targets in bone disorders like osteoporosis and osteogenesis imperfecta (Ansari et al. 2021; Tamura et al. 2023). In addition, MVs can be valuable when developing biomaterials to regenerate bone and engineer tissues (Han et al. 2024; Trentini et al. 2024). Moreover, a deeper understanding of the roles MVs have in vascular calcification may offer insights into the mechanisms underlying cardiovascular diseases (Welsh et al. 2024; Tamura et al. 2023; Li et al. 2022).

Although MVs have promising applications, conducting research into them is difficult particularly because various methodologies are available to isolate and characterize them. Among all the types of EVs, MVs have the unique ability to bind to collagen fibers. Therefore, digestion with collagenase is critical when isolating MVs from tissue or cell cultures. This enzymatic process breaks down collagen fibers, allowing MVs embedded within the ECM to be released, thereby enhancing the purity and yield of isolated vesicles (Boyan et al. 2022; Mizukami et al. 2023). Research has suggested that digestion with collagenase not only increases the yield of isolated MVs but also improves their mineralization potential by selecting vesicles enriched with apatite and presenting higher ALP activity (Mizukami et al. 2023; Sebinelli et al. 2022; Cruz et al. 2020). This makes digestion with collagenase a valuable tool to isolate MVs within higher mineralization capacity. Additionally, traditional isolation techniques (e.g., differential centrifugation, ultracentrifugation, filtration, and density gradient centrifugation) can be employed to recover isolated MVs after digestion with collagenase. Nevertheless, variability in protocols, equipment, and parameters can impact the yield and purity of MVs and reproducibility across studies (Welsh et al. 2024; Wakker et al. 2023). Similarly, current characterization methods, including the determination of ALP activity, TEM, and proteomic assays, provide only partial insight into the morphology, size, and composition of MVs. To date, it has been impossible to capture a complete picture of the properties of MVs by using a single technique (Crescitelli et al. 2021; Shami-shah et al. 2023). The lack of standardized characterization approaches complicates comparisons across studies, often resulting in discrepant findings regarding the characteristics, biological functions, and therapeutic potential of MVs (Yang et al. 2024; Zhuang et al. 2024).

In this systematic review, we have assessed current knowledge on MVs released by digestion with collagenase. We aimed to identify key limitations and propose practical guidelines on how to standardize research into MVs. This review also serves as a guide to where we stand on research into MVs, sheds light on the future of the field, and identifies which type of information is missing or remains underexplored. Beyond this systematic review, we have created a comprehensive digital library of MVs that will serve as a resource to advance the field, enable robust comparisons, and innovate research into MVs. We have constructed the digital library aiming to make it interactive and open to researchers. The digital library is dynamic and allows users to apply filters and taxonomy to find and interconnect the data resulting from this review.

Materials and methods

For this systematic review, the updated guidelines of Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) were followed (Page et al. 2021). The primary research question was “What is the state-of-the-art situation regarding research into isolation, characterization methods, and composition of collagenase-released MVs?” The question was formulated according to the Population, Interventions, Control, and Outcomes (PICO) framework. In the specific case of this review, Population (P) refers to MVs released by using collagenase, Intervention (I) refers to isolation and characterization techniques and compositional analysis, Comparison (C) corresponds to different types of collagenase enzymes (e.g., type I, type II) or additional enzyme techniques, and Outcome (O) comprises comprehensive insights into state-of-the-art methods and composition, aiming at standardizing or innovating approaches.

Eligibility criteria

The inclusion criteria were research studies only, experimental studies using either cellular or animal models, intervention studies examining MVs released by treatment with collagenase or collagenase combined with another digestive enzyme, presence of at least one characterization method, evaluation of composition-related parameters, studies with statistical analysis, and studies published in English. The exclusion criteria were non-research articles such as letters to the editor, commentaries, reviews, interviews, case reports/case series, qualitative studies, and systematic reviews or studies reporting non-collagenase-released MVs.

Literature search protocol and data extraction

A comprehensive literature search was conducted by including the international PROSPERO and Cochrane systematic review registries, but this search did not retrieve any registered reviews summarizing methodologies for isolating and characterizing MVs. A systematic search of databases such as PubMed (National Library of Medicine, Bethesda), Scopus, and Web of Knowledge from January 2000 through December 2023 was conducted. The search terms included “Collagenase-released matrix vesicles,” “Matrix vesicles,” and “Mineralizing vesicles” searched in a full-text body. After duplicates were removed, two reviewers (Chaves Filho and Ramos) independently screened titles and abstracts to reduce potential bias. Full-text articles that met the initial screening criteria were then evaluated against the eligibility criteria, and the final selection resulted from a consensus reached by the two reviewers.

To facilitate analysis, quantitative data concerning the comparison of ALP activity between cell-originated MVs and released MVs and between naturally occurring MVs and stimulated MVs were extracted as numerical values from the text and figures of each study and are expressed as relative values compared to the respective control group.

Digital library

The digital library was built by employing Tainacan (https://tainacan.org), a free software used to create digital repositories from the characterization of a metadata structure selected to represent the studies included in this systematic review. The library is available online at http://200.144.245.117/mineralizing-extracellular-vesicles. All the metadata structure was agreed on by all the authors of this review. Figure 1 depicts an example of a selection that can be executed by the users. The search can be filtered by tissue origin, study time, cell lineage, and authors. Different metadata like authors, country, journal tittle, doi, year of publication, and tissue origin, among others, can be selected and sorted for exhibition after filtering.

Fig. 1.

Example of the exhibition of the digital library of mineralizing vesicles. Different filters like tissue origin, study time, cell lineage, and authors can be applied, and several metadata including authors, country, journal title, DOI, year of publication, and tissue origin, among others, can be selected and sorted for exhibition after filtering

Results

Study selection

Initially, we identified 1385 potential studies. After we removed duplicates, 761 studies remained. Then, we reviewed the Abstracts for relevance and excluded 581 studies that did not address the primary research question. After that, we assessed 180 full-text articles and excluded 109 studies. Finally, 71 studies met the inclusion criteria, and we selected them for this systematic review, as illustrated in Fig. 2.

Fig. 2.

Flowchart of the process used to select studies for this systematic review. The PRISMA guidelines were followed

General characteristics of the studies included in the review

Table 1 summarizes the general characteristics of the selected studies, which employed various biological models to investigate the release and characteristics of MVs isolated from different cell lines derived from a range of tissues. Most studies (38 studies) (Sebinelli et al. 2022; Cruz et al. 2020; Sekrecka-Belniak et al. 2010; D'Angelo et al. 2001; Yan et al. 2022; Wiesmann et al. 2003; Wu et al. 2002, 2003; Damek-Poprawa et al. 2006; Balcerzak et al. 2007, 2008; Jaovisidha et al. 2002; Li et al. 2008, 2010; Ren et al. 2015; Bechkoff et al. 2008; Abdallah et al. 2014; Kirsch et al. 2003, 2000a, 2000b; Roberts et al. 2007; Solomon et al. 2007; Thouverey et al. 2009a; Minashima et al. 2012; Veschi et al. 2020; Nahar et al. 2008; Yi et al. 2022; Garimella et al. 2006, 2004a, 2004b; Maki et al. 2000; Guicheux et al. 2000; Zhang et al. 2005; Stewart et al. 2006; Plaut et al. 2019; Wang and Kirsch 2002; Genge et al. 2003; Bessueille et al. 2020) focused on cartilage, followed by bone (18 studies) (Mizukami et al. 2023; Sebinelli et al. 2022; Woeckel et al. 2010; Sawada et al. 2007; Jiang et al. 2013; You et al. 2020; Liu et al. 2014; Arivalagan et al. 2023; Zhou et al. 2013; Li et al. 2019, 2016; Cmoch et al. 2011; Thouverey et al. 2009b, 2011; Yamada et al. 2003; Simão et al. 2010; Hessle et al. 2002; Kawakubo et al. 2011), muscle (11 studies) (Chen et al. 2008, 2018, 2021, 2010a, 2010b; Roszkowska et al. 2018; Chaturvedi et al. 2015; Hsu et al. 2000a, 2000b; Lee et al. 2014; Hsu and Abbo 2004), dentin (two studies) (Kuzynski et al. 2014; Chaudhary et al. 2016), and connective (one study) (Mora-Navarro et al. 2020) and mesenchymal tissues (one study) (Yi et al. 2022). Tissue samples were sourced from animals, including chickens (Sebinelli et al. 2022; Cruz et al. 2020; Sekrecka-Belniak et al. 2010; D'Angelo et al. 2001; Wiesmann et al. 2003; Wu et al. 2002, 2003; Damek-Poprawa et al. 2006; Balcerzak et al. 2007, 2008; Li et al. 2008, 2010; Ren et al. 2015; Bechkoff et al. 2008; Abdallah et al. 2014; Kirsch et al. 2003, 2000a, 2000b; Roberts et al. 2007; Thouverey et al. 2009a; Veschi et al. 2020; Zhang et al. 2005; Stewart et al. 2006; Plaut et al. 2019; Wang and Kirsch 2002; Genge et al. 2003), mice (Yan et al. 2022; Minashima et al. 2012; Nahar et al. 2008; Garimella et al. 2006, 2004a, 2004b; Guicheux et al. 2000; Bessueille et al. 2020; Chaturvedi et al. 2015; Chen et al. 2018, 2021; Lee et al. 2014; Kuzynski et al. 2014; Chaudhary et al. 2016), humans (Yi et al. 2022; Woeckel et al. 2010; Jiang et al. 2013; Liu et al. 2014; Zhou et al. 2013; Li et al. 2019, 2016; Cmoch et al. 2011; Thouverey et al. 2009b, 2011; Yamada et al. 2003; Kawakubo et al. 2011; Mora-Navarro et al. 2020), mouse (Mizukami et al. 2023; Sebinelli et al. 2022; Sawada et al. 2007; You et al. 2020; Simão et al. 2010; Hessle et al. 2002; Roszkowska et al. 2018), cows and oxen (Solomon et al. 2007; Maki et al. 2000; Genge et al. 2003; Chen et al. 2008, 2010a, 2010b), rabbits (Hsu et al. 2000a, 2000b; Hsu and Abbo 2004), and pigs (Jaovisidha et al. 2002) (number of studies = 26, 14, 14, 7, 6, 3, and 1, respectively).

Table 1.

General characteristics of the studies included in the qualitative analysis

Reference	Tissue origin	Animal origin	Type of cell-related	Lineage	Stimuli or model to MVs production	Time for production	Collagenase type	Enzyme []	Method	MVs characterization	MVs composition	Diameter (nm)	Genetic material	Relative increase in ALP activity (%; cell x MVs)	Relative increase in ALP activity (%; natural x-stimulated MVs)
Mizukami et al. 2023)	Bone	Mouse	Osteoblast	Primary osteoblast	BMP2 and B-glycerophosphate	14 d	1	500 U/mL	Collagenase + trypsin + centrifugation	ALP, NTA, western blot, and gene expression	ALP, annexin 5, and CD9	< 200	+	850	N.P
Mizukami et al. 2023)	Bone	Mouse	Osteoblast	Primary osteoblast	BMP2	14 d	1	500 U/mL	Collagenase + trypsin + centrifugation	ALP, NTA, western blot, and gene expression	ALP, annexin 5, and CD9	< 200	+	850	N.P
Arivalagan et al. 2023)	Bone	Mouse	Osteoblast	MC3T3-TRIP-1OE	N.A	14 d	N.I	1.5 mg/mL	Collagenase + centrifugation	TEM, NTA, and proteomics	289 proteins	< 200	+	N.P	N.P
Arivalagan et al. 2023)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	N.A	N.A	1	300 U/g of tissue	Collagenase + centrifugation	ALP, AFM, and DLS	ALP	> 200	-	173	N.P
Yan et al. 2022)	Cartilage	Mice	Chondrocyte	Primary chondrocyte	Osteoarthritis	2 d	D	N.I	Collagenase + deoxyribonuclease I + centrifugation	TEM, NTA, western blot, and flow cytometry	Alix, CD9, LC3B, and Tsg101	> 200	-	N.P	N.P
Yi et al. 2022)	Mesenchymal	Human	Dental follicle cells	Primary follicle cell	Osteogenic inducers	7 d	1A	300 U/mL	Collagenase + centrifugation	TEM, NTA, X-ray, and western blot	ALP, CD63, TSG101, HSP70, and Ca and phosphate	> 200	-	N.P	N.P
Chen et al. 2021)	Muscular	Mice	VSMCS	Primary VSMCS	Osteogenic inducers	N.I	1A	500 U/mL	Collagenase + centrifugation	Western blot	Annexin (2, 4, and 10), CD9, CD63, and CD 81	N.I	-	NP	NP
Bessueille et al. 2020)	Cartilage	Mice	Chondrocyte	Primary chondrocyte	SiRNA 1 and 2	N.A	1	200 U/mL	Collagenase + centrifugation	Western blot	ALP	N.I	-	N.P	N.P
Cruz et al. 2020)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	N.A	N.A	1	300 U/g of tissue	Collagenase + centrifugation	ALP, DLS, and AFM	ALP	< 200	-	266	N.P
Mora-Navarro et al. 2020)	Conjunctive	Human	Fibroblast	HVOX	N.A	2 d	2	N.I	Collagenase + centrifugation	TEM and NTA	N.I	> 200	-	N.P	N.P
Veschi et al. 2020)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	N.A	N.A	1	200 U/g of tissue	Collagenase + centrifugation	Western blot, protein, and cholesterol content	ALP, annexin 6, cholesterol, and lipids	N.I	-	N.P	N.P
You et al. 2020)	Bone	Mouse	Osteoblast	MC3T3-E1	LPS	7 d	1A	2.5 mg/mL	Collagenase + centrifugation	TEM, AFM, and western blot	ALP, phospo1, Npp1, CD9, and CD63	N.I	-	N.P	N.P
Li et al. 2019)	Bone	Human	Osteoblast	SaO2	Osteogenic inducers	14 d	1A	200 U/mL	Collagenase + centrifugation	ALP and TEM	ALP and mineralization	< 200	-	N.P	N.P
Plaut et al. 2019)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	N.A	N.A	1A	200 U/g of tissue	Collagenase + centrifugation	ALP, DLS, AFM, and TEM	ALP	< 200	-	N.P	140
Chen et al. 2018)	Muscular	Mice	VSMCS	Primary VSMCS	Osteogenic inducers	N.I	1A	500 U/mL	Collagenase + centrifugation	TEM and western blot	Annexin (2,5 and 9), CD9, CD63 and CD 81	< 200	-	N.P	N.P
Roszkowska et al. 2018)	Muscular	Mouse	VSMCS	MOVAS-1	Osteogenic inducers	21 d	1A	0.5 mg/mL	Collagenase + centrifugation	ALP and TEM	ALP	< 200	-	N.P	160
Chaudhary et al. 2016)	Dentin	Mice	Pre-odontoblast	17IIA11	Osteogenic inducers	3 d	1A	2.5 mg/mL	Collagenase + centrifugation	NTA, TEM, cryo-electron microscopy, and western blot	ALP, annexin 5, PHOSPHO1, LAMP 1, and H3A	< 200	-	N.P	100
Li et al. 2016)	Bone	Human	Osteoblast	SaOS2	SiRNA 1 and 2	N.I	1A	1 mg/mL	Collagenase + centrifugation	Western blot, gene expression, and mineralization	ALP and annexins (2, 5, and 6)	N.I	+	N.P	−50
Chaturvedi et al. 2015)	Muscular	Mice	VSMCS	Primary VSMCS	N.A	N.A	1A	500 U/mL	Collagenase + centrifugation	Gene expression	miRNA	N.I	+	N.P	N.P
Ren et al. 2015)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	N.A	N.A	1A	500 U/g of tissue	Collagenase + centrifugation	Infrared	Phosphate substrates	> 200	-	N.P	N.P
Kuzynski et al. 2014)	Dentin	Mice	Pre-odontoblast	17IIA11	Osteogenic inducers	9 d	1A	2.5 mg/mL	Collagenase + centrifugation	AFM	N.I	< 200	-	N.P	N.P
Abdallah et al. 2014)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	N.A	N.A	1	200 U/g of tissue	Collagenase + centrifugation	ALP and lipids	ALP and lipids	N.I	-	507	N.P
Liu et al. 2014)	Bone	Human	Osteoblast	Saos-2	Osteogenic inducers	9 d	1A	1 mg/mL	Collagenase + centrifugation	ALP, TEM, western blot, and mineralization	ALP, CD9, and CD63	> 200	-	N.P	250
Liu et al. 2014)	Bone	Human	Osteoblast	Saos-2	Osteogenic inducers and FGF	9 d	1A	1 mg/mL	Collagenase + centrifugation	ALP, TEM, western blot, and mineralization	ALP, CD9, and CD63	> 200	-	N.P	100
Lee et al. 2014)	Muscular	Mice	VSMCS	Primary VSMCS	Osteogenic inducers	8 d	1A	N.I	Collagenase + centrifugation	ALP	ALP	N.I	-	N.P	N.P
Jiang et al. 2013)	Bone	Human	Osteoblast	Saos-2	Osteogenic inducers	7 d	N.I	1 mg/mL	Collagenase + centrifugation	TEM, western blot, and mass spectrometry	Protein kinase C and Ral-A	< 200	-	N.P	N.P
Jiang et al. 2013)	Bone	Human	Osteoblast	U2-OS	Osteogenic inducers	7 d	N.I	1 mg/mL	Collagenase + centrifugation	TEM, western blot, and mass spectrometry	Protein Kinase C and Ral-A	< 200	-	N.P	N.P
Zhou et al. 2013)	Bone	Human	Osteoblast	Saos-2	Osteogenic inducers	6 d	1A	1 mg/mL	Collagenase + centrifugation	ALP, TEM, mineralization, and mass spectrometry	ALP, CD9, CD63, HSP70, calnexin, calreticulin, S100A10, and b-actin	> 200	-	N.P	210
Minashima et al. 2012)	Cartilage	Mice	Chondrocyte	Primary chondrocyte	N.A	14 d	1A	500 U/mL	Collagenase + trypsin + centrifugation	Infrared and immunoblot	Annexins (2, 5, and 6) and Ca and phosphate	N.I	-	N.P	N.P
Minashima et al. 2012)	Bone	Human	Osteoblast	Saos-2	Osteogenic inducers	7 d	1A	100 U/mL	Collagenase + centrifugation	ALP, TEM, mineralization, and immunoblot	ALP, annexins (1, 2, 4, 5, 6, and 7) Hsc70 and b-actin	< 200	-	N.P	N.P
Thouverey et al. 2011)	Bone	Human	Osteoblast	Saos-2	Osteogenic inducers	7 d	1A	100 U/mL	Collagenase + centrifugation	ALP, mass spectrometry, and proteomics	ALP, proteins, Ca, and phosphate	> 200	-	1439	N.P
Kawakubo et al. 2011)	Bone	Human	Osteoblast	NOS-1	ZnSO4	5 d	N.I	580 U/mL	Collagenase + centrifugation	TEM and x-ray	protein, Ca, phosphate, and zinc	> 200	-	N.P	N.P
Woeckel et al. 2010)	Bone	Human	Osteoblast	SV-HFO	D vitamin	10 d	N.I	1 mg/mL	Collagenase + dispase + centrifugation	ALP, TEM, and flow cytometry	ALP	N.I	-	N.P	175
Sekrecka-Belniak et al. 2010)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	N.A	N.A	1A	200 U g/of tissue	Collagenase + trypsin and centrifugation	ALP and enzyme activities	ALP	N.I	-	N.P	N.P
Simão et al. 2010)	Bone	Mouse	Osteoblast	Primary osteoblast	Osteogenic inducers	18 d	N.I	0.45%	Collagenase + centrifugation	ALP, TEM, and enzyme assays	ALP	N.I	-	N.P	N.P
Chen et al. 2010a)	Muscular	Bovine	VSMCs	Primary VSMCs	Osteogenic inducers	7 d	1A	500 U/mL	Collagenase + centrifugation	ALP and mineralization	ALP	N.I	-	N.P	400
Li et al. 2010)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	N.A	N.A	1A	200 U g/of tissue	Collagenase + centrifugation	ALP, mineralization, and electrophoreses	ALP	< 200	-	N.P	N.P
Chen et al. 2010b)	Muscular	Bovine	VSMCS	Primary VSMCS	Osteogenic inducers	7 d	1A	500 U/mL	Collagenase + centrifugation	ALP and mineralization	ALP	N.I	-	N.P	455
Chen et al. 2010b)	Muscular	Bovine	VSMCS	Primary VSMCS	Osteogenic inducers and verapamil	7 d	1A	500 U/mL	Collagenase + centrifugation	ALP and mineralization	ALP	N.I	-	N.P	455
Thouverey et al. 2009a)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	N.A	N.A	1A	500 U/g of tissue	Collagenase + centrifugation	ALP, TEM infrared, mineralization, and enzyme activities	ALP and Caveolin 1	> 200	-	212,6	N.P
Thouverey et al. 2009b)	Bone	Human	Osteoblast	SaOS2	Osteogenic inducers	9 d	1A	100 U/mL	Collagenase + centrifugation	ALP, TEM, immunoblot, mineralization, and lipids	ALP, annexins (2 and 6) cholesterol, and lipids	> 200	-	1438	N.P
Chen et al. 2008)	Muscular	Bovine	VSMCS	Primary VSMCS	Osteogenic inducers	7 d	1A	500 U/mL	Collagenase + centrifugation	ALP, western blot, and mineralization	ALP and annexins (2 and 4)	N.I	-	N.P	234
Li et al. 2008)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	N.A	N.A	N.I	200 U/g of tissue	Collagenase + centrifugation	Mineralization, infrared, and x-ray	Ca and phosphate	N.I	-	N.P	N.P
Bechkoff et al. 2008)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	N.A	N.A	N.I	200 U/g of tissue	Collagenase + centrifugation	ALP, mineralization, and electrophoreses	ALP, Ca, and phosphate	< 200	-	N.P	N.P
Nahar et al. 2008)	Cartilage	Mice	Chondrocyte	Primary chondrocyte	N.A	28 d	N.I	1000 U/mL	Collagenase + centrifugation	Western blot	ALP and BMP (2, 4, 5, and 6)	N.I	-	N.P	N.P
Balcerzak et al. 2008)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	N.A	N.A	1	200 U/g of tissue	Collagenase + centrifugation	TEM, western blot, mineralization, and proteomics	ALP and ECM-derivates	< 200	-	N.P	N.P
Balcerzak et al. 2007)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	N.A	N.A	1	200 U/g of tissue	Collagenase + centrifugation	ALP, TEM, lipids and proteins, and mineralization	ALP, cholesterol, and lipids	< 200	-	N.P	N.P
Sawada et al. 2007)	Bone	Mouse	Osteoblast	MC3T3-E1	Osteogenic inducers	17 d	N.I	500 U/mL	Collagenase + centrifugation	ALP, TEM, western blot, gene expression, and lipids	ALP, annexins (2 and 5), osteocalcin, lipids, and caveolin	N.I	+	500	N.P
Roberts et al. 2007)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	Osteogenic inducers	12 d	N.I	0.45%	Collagenase + centrifugation	ALP and western blot	ALP and phospho1	N.I	-	N.P	400
Damek-Poprawa et al. 2006)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	N.A	N.A	N.I	0.50%	Hyaluronidase + collagenase + centrifugation	ALP and lipids	ALP and lipids	N.I	-	210	N.P
Solomon et al. 2007)	Cartilage	Bovine	Chondrocyte	Primary chondrocyte	N.A	N.A	1	1 mg/mL	Collagenase + centrifugation	Phosphate	Phosphate	N.I	-	N.P	N.P
Garimella et al. 2006)	Cartilage	Mice	Chondrocyte	Primary chondrocyte	N.A	N.I	N.I	1000 U/mL	Collagenase + centrifugation	Mineralization and infrared	Phosphate substrates	N.I	-	N.P	N.P
Balcerzak et al. 2006)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	N.A	N.A	1	200 U/g of tissue	Collagenase + trypsin + centrifugation	Mineralization and lipids	Ca and lipids	N.I	-	N.P	N.P
Stewart et al. 2006)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	N.A	10 d	N.I	0.70%	Collagenase + trypsin + centrifugation	Immunoblot	Phospho 1	N.I	-	N.P	300
Zhang et al. 2005)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	N.A	N.A	1	200 U/mL	Collagenase + trypsin + centrifugation	ALP, TEM, immunoblot, and infrared	ALP, Ca, and phosphate	< 200	-	N.P	N.P
Garimella et al. 2004b)	Cartilage	Mice	Chondrocyte	Primary chondrocyte	N.A	N.A	N.I	0.45%	Collagenase + centrifugation	ALP and mineralization	ALP, Ca, and phosphate	N.I	-	N.P	N.P
Garimella et al. 2004a)	Cartilage	Mice	Chondrocyte	Primary chondrocyte	N.A	17 d	N.I	2.5 mg/mL	Collagenase + centrifugation	ALP and mineralization	ALP, Ca, and phosphate	N.I	-	1000	N.P
Hsu and Abbo 2004)	Muscular	Rabbit	VSMCS	Primary VSMCS	Obesity model	N.A	1B	1.5 mg/mL	Collagenase + centrifugation	Mineralization	Ca and phosphate	N.I	-	N.P	N.P
Wiesmann et al. 2003)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	Retinoic acid	3 d	1A	500 U/mL	Collagenase + trypsin + centrifugation	Immunoblot and mineralization	Annexins (2,5 and 6), Ca, and phosphate	N.I	-	N.P	N.P
Wu et al. 2003)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	N.A	N.A	N.I	200 U/g of tissue	Collagenase + trypsin + centrifugation	Mineralization	Ca and phosphate	N.I	-	N.P	N.P
Kirsch et al. 2003)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	Retinoic acid	3 d	1A	500 U/g of tissue	Collagenase + centrifugation	ALP, TEM, and immunoblot	ALP and annexins (2, 5, and 6)	> 200	-	500	N.P
Yamada et al. 2003)	Bone	Human	Osteoblast	NOS-1	Chitosan	7 d	N.I	1000 U/mL	Collagenase + centrifugation	ALP e mineralization	ALP, Ca, and phosphate	< 200	-	N.P	155,5
Genge et al. 2003)	Cartilage	Bovine	Chondrocyte	Primary chondrocyte	N.A	N.A	1A	200 U/mL	Collagenase + trypsin + centrifugation	Lipids and mineralization	Lipids and Ca and phosphate	N.I	-	N.P	N.P
Genge et al. 2003)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	N.A	N.A	1A	200 U/mL	Collagenase + trypsin + centrifugation	Lipids and mineralization	Lipids and Ca and phosphate	N.I	-	N.P	N.P
Wu et al. 2002)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	N.A	N.A	1A	200 U/mL	Collagenase + trypsin + centrifugation	TEM, lipids, and mineralization	Lipids	N.I	-	N.P	N.P
Jaovisidha et al. 2002)	Cartilage	Pig	Chondrocyte	Primary chondrocyte	Nitroprusside	N.A	2	0.20%	Collagenase + centrifugation	TEM, ALP e mineralization	ALP	< 200	-	64	N.P
Wang and Kirsch 2002)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	Retinoic acid	3 d	1A	500 U/mL	Collagenase + centrifugation	ALP, immunoblot, and mineralization	ALP and annexins (2, 5, and 6)	N.I	-	960	N.P
Hessle et al. 2002)	Bone	Mouse	Osteoblast	Primary osteoblast	Osteogenic inducers	3 d	N.I	N.I	Collagenase + centrifugation	TEM and mineralization	Ca and phosphate	< 200	-	N.P	N.P
D'Angelo et al. 2001)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	N.A	N.A	1A	200 U/mL	Collagenase + trypsin + centrifugation	ALP, TEM, and immunoblot	ALP, annexins (2, 5, and 6), metalloproteinases, and TGF-beta	> 200	-	N.P	N.P
Maki et al. 2000)	Cartilage	Bovine	Chondrocyte	Primary chondrocyte	N.A	N.A	N.I	1000 U/mL	Collagenase + centrifugation	ALP, protein pattern, and enzyme assays	ALP	N.I	-	N.P	N.P
Guicheux et al. 2000)	Cartilage	Mice	Chondrocyte	ATDC5	Osteogenic inducers	21 d	2	500 U/mL	Collagenase + centrifugation	ALP and protein	ALP and collagen	N.I	-	1690	N.P
Hsu et al. 2000a)	Muscular	Habit	VSMCS	Primary VSMCS	Cholesterol and peanut oil	N.A	1B	0.10%	Collagenase + centrifugation	ALP, TEM, mineralization, and infrared	ALP, Ca, and phosphate	N.I	-	N.P	N.P
Hsu et al. 2000b)	Muscular	Rabbitt	VSMCS	Primary VSMCS	Cholesterol and peanut oil	N.A	N.I	0.10%	Collagenase + centrifugation	ALP, TEM, and Infrared	ALP, Ca, and phosphate	> 200	-	N.P	N.P
Kirsch and Claassen 2000)	Cartilage	Human	Chondrocyte	Primary chondrocyte	N.A	N.A	1A	200 U/of tissue	Collagenase + trypsin + centrifugation	ALP, TEM, and mineralization	ALP, Ca, and phosphate	< 200	-	N.P	N.P
Kirsch et al. 2000a)	Cartilage	Chicken	Chondrocyte	Primary chondrocyte	Retinoic acid and Zn	6 d	N.I	500 U/mL	Collagenase + trypsin + centrifugation	ALP and immunoblot	ALP and annexins (2, 5, and 6)	N.I	-	N.P	N.P
Kirsch et al. 2000b)	Bone	Mouse	Osteoblast	Primary osteoblast	BMP2 and B-glycerophosphate	14 d	1	500 U/mL	Collagenase + trypsin + centrifugation	ALP, NTA, western blot, and gene expression	ALP, annexin 5, and CD9	< 200	+	850	N.P

- not investigated; N.A. not applicable, N.I. not informed, N.P. not performed, Zn zinc, Ca calcium, ALP alkaline phosphatase, BMP2 bone morphogenetic protein-2, TEM transmittance electron microscopy, AFM atomic force microscopy, NTA nanoparticle tracking analysis, SiRNA small interfering ribonucleic acid NA, LPS lipopolysaccharide, ZnSO₄ zinc sulfate, VSMC vascular smooth muscular cells, d days, DLS dynamic light scattering, miRNA micro RNA, CD cluster of differentiation, Npp1 ectonucleotide pyrophosphatase/phosphodiesterase 1, TGF-beta transforming growth factor-beta

Regarding the cell from which MVs originated, chondrocytes (40 studies) (Sebinelli et al. 2022; Cruz et al. 2020; Sekrecka-Belniak et al. 2010; D'Angelo et al. 2001; Yan et al. 2022; Wiesmann et al. 2003; Wu et al. 2002, 2003; Damek-Poprawa et al. 2006; Balcerzak et al. 2007, 2008, 2006; Jaovisidha et al. 2002; Li et al. 2008, 2010; Ren et al. 2015; Bechkoff et al. 2008; Abdallah et al. 2014; Kirsch et al. 2003, 2000a, 2000b; Roberts et al. 2007; Solomon et al. 2007; Thouverey et al. 2009a; Minashima et al. 2012; Veschi et al. 2020; Nahar et al. 2008; Yi et al. 2022; Garimella et al. 2006, 2004a, 2004b; Maki et al. 2000; Guicheux et al. 2000; Zhang et al. 2005; Stewart et al. 2006; Plaut et al. 2019; Wang and Kirsch 2002; Genge et al. 2003; Bessueille et al. 2020; Hsu et al. 2000b), osteoblasts (18 studies) (Mizukami et al. 2023; Woeckel et al. 2010; Sawada et al. 2007; Jiang et al. 2013; You et al. 2020; Liu et al. 2014; Arivalagan et al. 2023; Zhou et al. 2013; Li et al. 2019, 2016; Cmoch et al. 2011; Thouverey et al. 2009b, 2011; Yamada et al. 2003; Simão et al. 2010; Hessle et al. 2002; Kawakubo et al. 2011; Chaudhary et al. 2016), vascular smooth muscle cells (VSMC) (ten studies) (Chen et al. 2008, 2018, 2021, 2010a, 2010b; Roszkowska et al. 2018; Chaturvedi et al. 2015; Hsu et al. 2000a; Lee et al. 2014; Hsu and Abbo 2004), or primary follicle dental cells (Yi et al. 2022) and fibroblasts (Mora-Navarro et al. 2020) (one study each) were the main source of MVs. With respect to chondrocytes, 38 studies (Sebinelli et al. 2022; Cruz et al. 2020; Sekrecka-Belniak et al. 2010; D'Angelo et al. 2001; Yan et al. 2022; Wiesmann et al. 2003; Wu et al. 2002, 2003; Damek-Poprawa et al. 2006; Balcerzak et al. 2007, 2008; Jaovisidha et al. 2002; Li et al. 2008, 2010; Ren et al. 2015; Bechkoff et al. 2008; Abdallah et al. 2014; Kirsch et al. 2003, 2000a, 2000b; Roberts et al. 2007; Solomon et al. 2007; Thouverey et al. 2009a; Minashima et al. 2012; Veschi et al. 2020; Nahar et al. 2008; Yi et al. 2022; Garimella et al. 2006, 2004a, 2004b; Maki et al. 2000; Zhang et al. 2005; Stewart et al. 2006; Plaut et al. 2019; Wang and Kirsch 2002; Genge et al. 2003; Bessueille et al. 2020; Hsu and Abbo 2004) used primary chondrocytes, while one study (Guicheux et al. 2000) employed the murine ATDC5 chondrogenic cell line. Osteoblast-derived MVs involved human cell lines, i.e., human osteoblastic cell line immortalized by simian virus 40 (SV-HFO) (Woeckel et al. 2010), human osteosarcoma cell lines, as Saos2 (Jiang et al. 2013; Liu et al. 2014; Zhou et al. 2013; Li et al. 2019, 2016; Cmoch et al. 2011; Thouverey et al. 2009b, 2011), NOS-1 (Yamada et al. 2003; Kawakubo et al. 2011), and U2-OS (Jiang et al. 2013). Murine primary osteoblasts (Mizukami et al. 2023; Simão et al. 2010; Hessle et al. 2002) and the murine cell lines MC3T3-E1 (Sawada et al. 2007; You et al. 2020) and MC3T3-TIP1 (Sebinelli et al. 2022; Arivalagan et al. 2023) were also the subject of some studies included here. Primary VSMCS were the subject of nine studies (Chen et al. 2008, 2018, 2021, 2010a, 2010b; Chaturvedi et al. 2015; Hsu et al. 2000a, 2000b; Lee et al. 2014), whereas one study (Roszkowska et al. 2018) used the murine lineage MOVAS-1. Primary follicle dentin cells (Yi et al. 2022) and HVOX fibroblast cell lines (Mora-Navarro et al. 2020) were less frequently employed.

Concerning the stimuli used to produce MVs, 34 studies (Sebinelli et al. 2022; Cruz et al. 2020; Sekrecka-Belniak et al. 2010; D'Angelo et al. 2001; Wiesmann et al. 2003; Wu et al. 2002, 2003; Damek-Poprawa et al. 2006; Balcerzak et al. 2007, 2008; Li et al. 2008, 2010; Ren et al. 2015; Bechkoff et al. 2008; Abdallah et al. 2014; Roberts et al. 2007; Solomon et al. 2007; Thouverey et al. 2009a; Minashima et al. 2012; Veschi et al. 2020; Nahar et al. 2008; Yi et al. 2022; Garimella et al. 2006, 2004a, 2004b; Maki et al. 2000; Zhang et al. 2005; Plaut et al. 2019; Kirsch et al. 2000b; Bessueille et al. 2020; Arivalagan et al. 2023; Chaturvedi et al. 2015; Hsu et al. 2000a; Mora-Navarro et al. 2020) obtained MVs without applying any external stimuli to enhance their release, whilst 23 studies (Minashima et al. 2012; Yi et al. 2022; Guicheux et al. 2000; Stewart et al. 2006; Sawada et al. 2007; Jiang et al. 2013; Liu et al. 2014; Zhou et al. 2013; Li et al. 2019; Cmoch et al. 2011; Thouverey et al. 2009b, 2011; Simão et al. 2010; Hessle et al. 2002; Chen et al. 2008, 2018, 2021, 2010a, 2010b; Roszkowska et al. 2018; Lee et al. 2014; Kuzynski et al. 2014; Chaudhary et al. 2016) applied traditional osteogenic inducers (mainly beta-glycerophosphate or ascorbic acid). Additional stimuli included fibroblast growth factor 2 (FGF) (Liu et al. 2014), retinoic acid (Wiesmann et al. 2003; Kirsch et al. 2003, 2000a; Wang and Kirsch 2002), and retinoic acid combined with zinc (Kirsch et al. 2000a), along with other inducers such as bone morphogenetic protein-2 (BMP-2) (Mizukami et al. 2023), chitosan (Yamada et al. 2003), vitamin D (Woeckel et al. 2010), lipopolysaccharide (LPS) (You et al. 2020), and small interfering RNA (SiRNA) (Li et al. 2016). An obesity diet (Hsu and Abbo 2004) and a supplemented cholesterol plus peanut oil diet (Hsu et al. 2000b) models were also employed.

Thirty-eight studies (Sebinelli et al. 2022; Cruz et al. 2020; Sekrecka-Belniak et al. 2010; D'Angelo et al. 2001; Wu et al. 2002, 2003; Damek-Poprawa et al. 2006; Balcerzak et al. 2007, 2008; Jaovisidha et al. 2002; Li et al. 2008, 2010; Ren et al. 2015; Bechkoff et al. 2008; Abdallah et al. 2014; Roberts et al. 2007; Solomon et al. 2007; Thouverey et al. 2009a; Veschi et al. 2020; Yi et al. 2022; Maki et al. 2000; Zhang et al. 2005; Plaut et al. 2019; Kirsch et al. 2000b; Genge et al. 2003; Garimella et al. 2004b; Chaturvedi et al. 2015; Hsu et al. 2000a, 2000b; Chen et al. 2018; Hsu and Abbo 2004) did not specify the time needed to produce MVs and isolated them directly from tissue. In Table 1, this is indicated as not applicable (N.A.). Cultures lasted from 2 (Yan et al. 2022; Mora-Navarro et al. 2020) to 28 days (Nahar et al. 2008), but four studies (Wiesmann et al. 2003; Garimella et al. 2006; Li et al. 2016; Chen et al. 2021) failed to report this information.

The digestion of the ECM by enzymes, such as collagenase, is critical for isolating MVs. Thirty-four studies (Sekrecka-Belniak et al. 2010; D'Angelo et al. 2001; Wu et al. 2002; Ren et al. 2015; Kirsch et al. 2003, 2000b; Thouverey et al. 2009a, 2009b, 2011; Minashima et al. 2012; Yi et al. 2022; Plaut et al. 2019; Wang and Kirsch 2002; Genge et al. 2003; Li et al. 2010, 2019, 2016; You et al. 2020; Liu et al. 2014; Zhou et al. 2013; Cmoch et al. 2011; Chen et al. 2008, 2018, 2021, 2010a, 2010b; Roszkowska et al. 2018; Chaturvedi et al. 2015; Lee et al. 2014; Kuzynski et al. 2014; Chaudhary et al. 2016) used collagenase 1A, while 14 studies (Mizukami et al. 2023; Sebinelli et al. 2022; Cruz et al. 2020; Wiesmann et al. 2003; Balcerzak et al. 2007, 2008, 2006; Abdallah et al. 2014; Solomon et al. 2007; Veschi et al. 2020; Zhang et al. 2005; Bessueille et al. 2020; Arivalagan et al. 2023; Chen et al. 2021) employed collagenase 1. Collagenase 1b (2 studies) (Hsu et al. 2000a; Hsu and Abbo 2004), collagenase 2 (3 studies) (Jaovisidha et al. 2002; Guicheux et al. 2000; Mora-Navarro et al. 2020), and collagenase D (1 study) (Yan et al. 2022) were also used. The predominant isolation method was digestion with collagenase combined with centrifugation, reported in 53 studies (Sebinelli et al. 2022; Cruz et al. 2020; Balcerzak et al. 2007, 2008; Jaovisidha et al. 2002; Li et al. 2008, 2010, 2019, 2016; Ren et al. 2015; Bechkoff et al. 2008; Abdallah et al. 2014; Kirsch et al. 2003; Roberts et al. 2007; Solomon et al. 2007; Thouverey et al. 2009a, 2009b, 2011; Veschi et al. 2020; Nahar et al. 2008; Garimella et al. 2006, 2004a, 2004b; Maki et al. 2000; Guicheux et al. 2000; Plaut et al. 2019; Wang and Kirsch 2002; Bessueille et al. 2020; Sawada et al. 2007; Jiang et al. 2013; You et al. 2020; Liu et al. 2014; Arivalagan et al. 2023; Zhou et al. 2013; Cmoch et al. 2011; Yamada et al. 2003; Simão et al. 2010; Hessle et al. 2002; Kawakubo et al. 2011; Chen et al. 2008, 2018, 2021, 2010a, 2010b; Roszkowska et al. 2018; Chaturvedi et al. 2015; Hsu et al. 2000a, 2000b; Lee et al. 2014; Hsu and Abbo 2004; Kuzynski et al. 2014; Chaudhary et al. 2016; Mora-Navarro et al. 2020), while 17 studies (Mizukami et al. 2023; Sekrecka-Belniak et al. 2010; D'Angelo et al. 2001; Wiesmann et al. 2003; Wu et al. 2002, 2003; Minashima et al. 2012; Yi et al. 2022; Zhang et al. 2005; Stewart et al. 2006; Kirsch et al. 2000a, 2000b; Genge et al. 2003; Balcerzak et al. 2006) applied collagenase combined with trypsin. Other studies used collagenase combined with dispase (Woeckel et al. 2010; Hsu et al. 2000a), hyaluronidase (Damek-Poprawa et al. 2006), or deoxyribonuclease (Yan et al. 2022). The concentration of collagenase and the respective unit differed across the studies. All the studies employed centrifugation approaches, but the set-up conditions differed.

Characterization and composition of MVs

The selected studies used several techniques, often in combination, to characterize MVs. The most common techniques were ALP detection or activity assays (45 studies) (Mizukami et al. 2023; Sebinelli et al. 2022; Cruz et al. 2020; Sekrecka-Belniak et al. 2010; D'Angelo et al. 2001; Damek-Poprawa et al. 2006; Balcerzak et al. 2007, 2008; Jaovisidha et al. 2002; Bechkoff et al. 2008; Abdallah et al. 2014; Kirsch et al. 2003, 2000a; Roberts et al. 2007; Thouverey et al. 2009a, 2009b, 2011; Veschi et al. 2020; Nahar et al. 2008; Yi et al. 2022; Maki et al. 2000; Zhang et al. 2005; Plaut et al. 2019; Wang and Kirsch 2002; Garimella et al. 2004b; Li et al. 2010, 2019, 2016; Woeckel et al. 2010; Sawada et al. 2007; You et al. 2020; Liu et al. 2014; Zhou et al. 2013; Cmoch et al. 2011; Yamada et al. 2003; Simão et al. 2010; Chen et al. 2008, 2021, 2010a, 2010b; Roszkowska et al. 2018; Hsu et al. 2000a, 2000b; Lee et al. 2014; Chaudhary et al. 2016) followed by TEM (30 studies) (Mizukami et al. 2023; D'Angelo et al. 2001; Yan et al. 2022; Wu et al. 2002; Jaovisidha et al. 2002; Kirsch et al. 2003, 2000b; Thouverey et al. 2009a, 2009b; Yi et al. 2022; Zhang et al. 2005; Balcerzak et al. 2008; Plaut et al. 2019; Woeckel et al. 2010; Sawada et al. 2007; Jiang et al. 2013; You et al. 2020; Liu et al. 2014; Zhou et al. 2013; Li et al. 2019; Cmoch et al. 2011; Simão et al. 2010; Hessle et al. 2002; Kawakubo et al. 2011; Roszkowska et al. 2018; Hsu et al. 2000a, 2000b; Chen et al. 2018; Chaudhary et al. 2016; Mora-Navarro et al. 2020) and mineralization assays (27 studies) (Wiesmann et al. 2003; Wu et al. 2002, 2003; Jaovisidha et al. 2002; Li et al. 2008, 2010, 2019, 2016; Bechkoff et al. 2008; Thouverey et al. 2009a, 2009b; Yi et al. 2022; Garimella et al. 2006, 2004b; Balcerzak et al. 2008; Wang and Kirsch 2002; Genge et al. 2003; Liu et al. 2014; Zhou et al. 2013; Cmoch et al. 2011; Yamada et al. 2003; Hessle et al. 2002; Chen et al. 2008, 2010a, 2010b; Hsu et al. 2000a; Hsu and Abbo 2004). Other techniques included AFM (five studies) (Sebinelli et al. 2022; Cruz et al. 2020; Plaut et al. 2019; You et al. 2020; Kuzynski et al. 2014), NTA (4 studies) (Mizukami et al. 2023; Yan et al. 2022; Yi et al. 2022; Chaudhary et al. 2016; Mora-Navarro et al. 2020), mass spectrometry (three studies) (Jiang et al. 2013; Zhou et al. 2013; Thouverey et al. 2011), and X-ray analysis (three studies) (Li et al. 2008; Yi et al. 2022; Kawakubo et al. 2011). Proteins were identified by immunoblotting (11 studies) (D'Angelo et al. 2001; Wiesmann et al. 2003; Kirsch et al. 2003, 2000a, 2000b; Minashima et al. 2012; Zhang et al. 2005; Stewart et al. 2006; Wang and Kirsch 2002; Cmoch et al. 2011; Thouverey et al. 2009b), western blot (16 studies) (Mizukami et al. 2023; Yan et al. 2022; Roberts et al. 2007; Veschi et al. 2020; Nahar et al. 2008; Yi et al. 2022; Balcerzak et al. 2008; Sawada et al. 2007; Jiang et al. 2013; You et al. 2020; Liu et al. 2014; Zhou et al. 2013; Chen et al. 2008, 2018, 2021; Chaudhary et al. 2016), or proteomics (three studies) (Balcerzak et al. 2008; Arivalagan et al. 2023; Thouverey et al. 2011). Eight studies (Wiesmann et al. 2003; Wu et al. 2002; Balcerzak et al. 2007; Abdallah et al. 2014; Veschi et al. 2020; Genge et al. 2003; Sawada et al. 2007; Thouverey et al. 2009b) evaluated the lipid content.

In terms of the composition of MVs, ALP was the most investigated component (Mizukami et al. 2023; Sebinelli et al. 2022; Cruz et al. 2020; Sekrecka-Belniak et al. 2010; D'Angelo et al. 2001; Damek-Poprawa et al. 2006; Balcerzak et al. 2007, 2008; Jaovisidha et al. 2002; Bechkoff et al. 2008; Abdallah et al. 2014; Kirsch et al. 2003, 2000a; Roberts et al. 2007; Thouverey et al. 2009a, 2009b, 2011; Veschi et al. 2020; Nahar et al. 2008; Yi et al. 2022; Maki et al. 2000; Zhang et al. 2005; Plaut et al. 2019; Wang and Kirsch 2002; Garimella et al. 2004b; Li et al. 2010, 2019, 2016; Woeckel et al. 2010; Sawada et al. 2007; You et al. 2020; Liu et al. 2014; Zhou et al. 2013; Cmoch et al. 2011; Yamada et al. 2003; Simão et al. 2010; Chen et al. 2008, 2021, 2010a, 2010b; Roszkowska et al. 2018; Hsu et al. 2000a, 2000b; Lee et al. 2014; Chaudhary et al. 2016), evaluated through activity assays and analysis of protein expression. Other proteins detected in MVs included annexins, with isoforms II (Mizukami et al. 2023; D'Angelo et al. 2001; Wiesmann et al. 2003; Kirsch et al. 2003, 2000a, 2000b; Minashima et al. 2012; Wang and Kirsch 2002; Sawada et al. 2007; Cmoch et al. 2011; Thouverey et al. 2009b; Li et al. 2016; Chen et al. 2008, 2018; Chaudhary et al. 2016), V (Mizukami et al. 2023; D'Angelo et al. 2001; Wiesmann et al. 2003; Kirsch et al. 2003, 2000a, 2000b; Minashima et al. 2012; Wang and Kirsch 2002; Sawada et al. 2007; Cmoch et al. 2011; Li et al. 2016; Chen et al. 2018; Chaudhary et al. 2016), and VI (D'Angelo et al. 2001; Wiesmann et al. 2003; Kirsch et al. 2003, 2000a, 2000b; Minashima et al. 2012; Veschi et al. 2020; Wang and Kirsch 2002; Cmoch et al. 2011; Thouverey et al. 2009b; Li et al. 2016) being the most cited (15, 13, and 11 studies respectively). Alix (Yan et al. 2022), BMP-2 (Nahar et al. 2008), caveolin-1 (Thouverey et al. 2009a; Sawada et al. 2007), collagen (Guicheux et al. 2000; Kirsch et al. 2000b), phospho-1 (Roberts et al. 2007; Stewart et al. 2006; You et al. 2020; Chaudhary et al. 2016), TGF-beta (D'Angelo et al. 2001), and metalloproteinases (D'Angelo et al. 2001), among other proteins (Veschi et al. 2020; Arivalagan et al. 2023; Thouverey et al. 2011; Kawakubo et al. 2011), were also reported. Fourteen studies reported Cluster of Differentiation (CD) markers, with CD9 (Mizukami et al. 2023; Yan et al. 2022; You et al. 2020; Liu et al. 2014; Zhou et al. 2013; Chen et al. 2018, 2021), CD63 (Yi et al. 2022; You et al. 2020; Liu et al. 2014; Zhou et al. 2013; Chen et al. 2018, 2021), and CD83 (Chen et al. 2018, 2021) being the most studied targets of MVs. Other components identified in MVs included ECM-derived proteins (Balcerzak et al. 2008), cholesterol (Balcerzak et al. 2007; Veschi et al. 2020; Thouverey et al. 2009b), calcium and phosphate (Wiesmann et al. 2003; Li et al. 2008; Bechkoff et al. 2008; Wu et al. 2003; Minashima et al. 2012; Yi et al. 2022; Zhang et al. 2005; Garimella et al. 2004a, 2004b; Genge et al. 2003; Yamada et al. 2003; Thouverey et al. 2011; Hessle et al. 2002; Kawakubo et al. 2011; Hsu et al. 2000a, 2000b; Hsu and Abbo 2004), phosphate substrates (Ren et al. 2015; Garimella et al. 2006), zinc (Kawakubo et al. 2011), and microRNAs (miRNAs) (Chaturvedi et al. 2015), among others.

Thirty-three studies reported size distribution measurements, with 19 studies (Mizukami et al. 2023; Cruz et al. 2020; Balcerzak et al. 2007, 2008; Jaovisidha et al. 2002; Bechkoff et al. 2008; Yi et al. 2022; Zhang et al. 2005; Plaut et al. 2019; Li et al. 2010; Jiang et al. 2013; Arivalagan et al. 2023; Cmoch et al. 2011; Yamada et al. 2003; Hessle et al. 2002; Roszkowska et al. 2018; Chen et al. 2018; Kuzynski et al. 2014; Chaudhary et al. 2016) reporting diameters smaller than 200 nm, and 14 studies (Sebinelli et al. 2022; D'Angelo et al. 2001; Yan et al. 2022; Ren et al. 2015; Kirsch et al. 2003, 2000b; Thouverey et al. 2009a, 2009b, 2011; Yi et al. 2022; Liu et al. 2014; Zhou et al. 2013; Li et al. 2019; Hsu et al. 2000b; Mora-Navarro et al. 2020) reporting particles larger than 200 nm. Interestingly, only five studies (Mizukami et al. 2023; Sawada et al. 2007; Arivalagan et al. 2023; Li et al. 2016; Chaturvedi et al. 2015) reported the genetic content of MVs, whereas the remaining studies focused on the protein and lipid content and inorganic composition of MVs.

ALP activity in MVSs

Thirty-two studies investigated ALP activity, a key marker of mineralization. Fourteen studies (Mizukami et al. 2023; Sebinelli et al. 2022; Cruz et al. 2020; Damek-Poprawa et al. 2006; Jaovisidha et al. 2002; Abdallah et al. 2014; Kirsch et al. 2003; Thouverey et al. 2009a, 2011; Guicheux et al. 2000; Garimella et al. 2004a; Wang and Kirsch 2002; Sawada et al. 2007; Cmoch et al. 2011) compared ALP activity in MVs and the mother cells, while 13 studies (Roberts et al. 2007; Stewart et al. 2006; Plaut et al. 2019; Woeckel et al. 2010; Liu et al. 2014; Zhou et al. 2013; Li et al. 2016; Hessle et al. 2002; Chen et al. 2008, 2010a, 2010b; Roszkowska et al. 2018; Chaudhary et al. 2016) compared ALP activity in naturally occurring and stimulated MVs. Compared to the cells from which MVs originated, MVs were enriched with ALP. As for stimulus, only two studies reported that the content of ALP did not increase in MVs isolated from cells treated with xenobiotics (Li et al. 2016; Chen et al. 2010b).

Discussion

In this systematic review, study selection and inclusion resulted in a final pool of 71 studies meeting the inclusion criteria. This highlights that it is important to select studies carefully, to ensure consistent findings across studies, particularly findings concerning the isolation and characterization of MVs released by using collagenase-based methodologies. Variability in isolation methods affects purity and remains a major challenge in research into MVs (Welsh et al. 2024; Shami-shah et al. 2023; Balcerzak et al. 2007; Hsu et al. 2000b), preventing reproducibility and comparability for downstream analysis. For instance, differences in enzyme concentrations, digestion durations, and centrifugation conditions have been shown to influence the yield and functional properties of MVs, as well as the presence of contaminants. Such differences hinder reliable comparisons across studies and emphasize the need for standardization. To address this challenge, we propose the following steps to achieve standardization:

1. Consensus guidelines: Multidisciplinary experts should establish standardized guidelines specifying enzyme type, concentration, digestion duration, and centrifugation parameters.

2. Reporting standards: Researchers should adopt detailed reporting practices, including enzyme lot numbers, preparation methods, and isolation conditions.

3. Systematic comparison studies: Pilot studies systematically comparing protocols under identical conditions are essential to evaluate their impact on MV´s purity, yield, and functionality.

Most studies included in this review employed collagenase 1A (31 studies) (Sekrecka-Belniak et al. 2010; D'Angelo et al. 2001; Wiesmann et al. 2003; Wu et al. 2002; Ren et al. 2015; Kirsch et al. 2003, 2000b; Thouverey et al. 2009a, 2009b, 2011; Minashima et al. 2012; Yi et al. 2022; Plaut et al. 2019; Wang and Kirsch 2002; Genge et al. 2003; Li et al. 2010, 2019, 2016; You et al. 2020; Liu et al. 2014; Zhou et al. 2013; Cmoch et al. 2011; Chen et al. 2008, 2018, 2021, 2010a, 2010b; Roszkowska et al. 2018; Chaturvedi et al. 2015; Lee et al. 2014; Kuzynski et al. 2014; Chaudhary et al. 2016), highlighting its effectiveness in digesting tissues for MV isolation. However, the variability in enzyme types (such as collagenase 1b or 2) and concentration underscores the need for unified protocols. Combining collagenase with other enzymes (such as trypsin or hyaluronidase as reported in 17 studies) could offer additional benefits by targeting residual protein connections or ECM components, improving MV yield and purity. Future studies should systematically evaluate such combinations to determine optimal conditions for MV isolation.

As shown in Table 1, the studies included in this review represent a broad set of biological models, types of cells, and tissues, underscoring the insertion of MVs in different physiological contexts (Sebinelli et al. 2022; Yan et al. 2022; Kirsch et al. 2000a; Arivalagan et al. 2023; Su et al. 2023). Varying the types of tissue, animal species, and cell lines used to isolate MVs introduces a considerable challenge when it comes to interpreting results across studies. On the other hand, such variability reveals an entire field for investigating MVs (Welsh et al. 2024).

A critical aspect of research into MVs lies in the isolation methods. Numerous studies (53; 74.66% of the studies included in this review) combined digestion with collagenase and centrifugation to isolate MVs. Collagenase is widely employed because it digests the ECM and releases MVs without affecting their integrity (Zhang et al. 2024; Boyan et al. 2022; Hsu et al. 2000b). This makes collagenase a valuable tool for isolating MVs for studies on bone mineralization (Yi et al. 2022) and vascular health (Li et al. 2022; Chen et al. 2021). However, the concentrations and respective units of collagenase used in the studies differed significantly. Standardizing digestion with collagenase in terms of the type of enzyme and digestion protocols could improve reproducibility and comparability across studies, advancing research into MVs (Zhang et al. 2024).

Collagenase digestion followed by centrifugation is the most reliable and used method for isolating MVs, as evidenced by its application in 53 studies. Nevertheless, variations in key parameters, including the type of collagenase, duration of digestion, and centrifugation conditions (speed, duration, and temperature), significantly influence the yield, purity, and functionality of the isolated MVs (Welsh et al. 2024; Buchet et al. 2013). These variations may impact the size distribution, concentration, and functional properties of MVs, potentially contributing to inconsistencies when different studies are compared (Sousa et al. 2023). Therefore, establishing standardized protocols that clearly specify the type of enzyme, time of digestion, and centrifugation conditions is essential to increase reproducibility and to advance research into MVs.

Characterization of MVs remains a cornerstone of understanding their functionality and composition. For example, detailed structural insights gained through TEM have directly advanced the understanding of MV-mediated mineralization by confirming the presence of electron-dense mineral deposits such as hydroxyapatite. This has reinforced their role in regenerative medicine and highlighted their potential as therapeutic carriers. ALP activity assays, TEM, and protein content analysis were the most employed methods, providing complementary insights into MV functionality, structure, and composition. ALP activity assays remain the gold standard for assessing mineralization potential (Zhang et al. 2024; Kirsch et al. 2000a; Bessueille et al. 2020; Andrilli et al. 2023), while TEM offers high-resolution visualization of MV morphology and mineral content. Protein analysis elucidates biochemical components essential for understanding MV-mediated mineralization. However, incorporating additional techniques such as NTA and mass spectrometry could enhance precision in size distribution, concentration measurements, and proteomic profiling, respectively. These advancements would provide a more comprehensive characterization of MVs, supporting their translational applications.

To characterize MVs as well as possible, ALP activity assays, TEM, and NTA should be combined (Welsh et al. 2024; Vaiaki and Falasca 2024). On the basis of high-resolution imaging, TEM provides detailed structural insights, allowing the size, shape, and membrane characteristics of MVs to be analyzed (Mizukami et al. 2023; Yan et al. 2022; Yi et al. 2022). TEM also confirms whether isolated particles are MVs (Mizukami et al. 2023; Yan et al. 2022; Mora-Navarro et al. 2020), given that it can reveal electron-dense mineral deposits, such as hydroxyapatite crystals, within or around MVs, underscoring the role of MVs in biomineralization (Mizukami et al. 2023; Yan et al. 2022). TEM paired with X-ray spectroscopy further validates the presence of calcium and phosphate in these deposits, offering a comprehensive evaluation of the structural and functional attributes of MVs (Yi et al. 2022; Kawakubo et al. 2011). NTA complements these findings by providing precise measurements of the concentration and size distribution of MVs through tracking of individual particles, enabling detailed population analyses (Mizukami et al. 2023; Yan et al. 2022; Chaudhary et al. 2016; Mora-Navarro et al. 2020). This set of data is crucial for standardizing and reliably characterizing MVs. By combining ALP assays, TEM, and NTA, researchers can achieve a balanced strategy to characterize MVs through integrated functional, structural, and morphological analyses. These complementary techniques collectively provide insights into the part played by MVs in mineralization (Cmoch et al. 2011; Chaudhary et al. 2016).

MVs are complex structures with diverse biochemical and structural compositions, as highlighted by the studies included here. Our qualitative analysis identified several key components within MVs, including proteins, lipids, and minerals. ALP, annexins, and osteocalcin are among the most cited proteins. Each protein has an essential part in mineralization, cellular signaling, and ECM remodeling (Zhuang et al. 2024; Han et al. 2024; Yan et al. 2022; Kirsch and Claassen 2000). Annexins, especially isoforms II, V, and VI, were the most extensively studied. These isoforms contribute to membrane stabilization, vesicle budding, and ion regulation, which is paramount for MVs to function properly during biomineralization processes (Kirsch et al. 2000a, 2000b; Wang and Kirsch 2002; Cmoch et al. 2011; Li et al. 2016; Su et al. 2023). Additionally, the identification of osteocalcin (Sawada et al. 2007), TGF-beta (D'Angelo et al. 2001), collagen (Guicheux et al. 2000; Kirsch et al. 2000b), and metalloproteinases (D'Angelo et al. 2001) within MVs highlights their dual structural and signaling roles. These proteins contribute to the formation of the mineralized ECM and regulate cellular differentiation, migration, and tissue remodeling (Boyan et al. 2022; Mizukami et al. 2023; Guicheux et al. 2000). The protein composition of MVs is therefore integral to their function as mediators of mineralization and facilitators of tissue repair (Anto et al. 2024).

The reviewed studies consistently reported that MVs are enriched with ALP compared to their parent cells. This enrichment suggests that MVs play a more pivotal role in mineralization and potentially serve as an effective carrier of ALP (Anto et al. 2024). Among the 14 studies (Mizukami et al. 2023; Sebinelli et al. 2022; Cruz et al. 2020; Damek-Poprawa et al. 2006; Jaovisidha et al. 2002; Abdallah et al. 2014; Kirsch et al. 2003; Thouverey et al. 2009a, 2011; Guicheux et al. 2000; Garimella et al. 2004a; Wang and Kirsch 2002; Cmoch et al. 2011) comparing ALP activity in cells and MVs isolated from them, the activity of the enzyme increased by between 173% (18) and 1690% (47) within MVs (the average increase was 722 ± 496%). Interestingly, several studies demonstrated that cells treated with osteogenic inducers (e.g., beta-glycerophosphate, ascorbic acid) regulate ALP activity in the MVs released by them. Of note, two studies reported that ALP activity did not increase after treatment. The first study used SiRNA to interfere in ALP gene expression, which specifically reduced ALP activity, as expected (Li et al. 2016). The second study combined osteogenic inducers with verapamil, a calcium channel blocker, which hindered ALP activity (Chen et al. 2010b).

In contrast, 12 studies reported increased ALP activity in MVs released under stimulation compared to non-stimulated conditions (Stewart et al. 2006; Plaut et al. 2019; Woeckel et al. 2010; Liu et al. 2014; Zhou et al. 2013; Thouverey et al. 2009b; Yamada et al. 2003; Chen et al. 2008, 2010a, 2010b; Roszkowska et al. 2018; Chaudhary et al. 2016). ALP activity in MVs increased by 216 ± 137%, on average. The highest relative increase (400%) was reported in primary VSMC treated with osteogenic inducers for 7 days. This same study showed that verapamil considerably inhibited the rise in ALP content in MVs (Chen et al. 2010b), underscoring the potential role of these vesicles in pathological calcification research, diagnosis, and therapeutic targeting. Overall, comparing ALP activity in MVs and their parent cells under stimulation or natural conditions provides a valuable tool to assess the mineralization potential of MVs and their roles in tissue regeneration and pathological calcification.

The presence of calcium and phosphate ions in MVs is crucial because these ions are essential for the formation of apatite minerals, the primary component of the inorganic part of bone (Yi et al. 2022; Li et al. 2019, 2016). When these ions accumulate within MVs, they serve as nucleation sites for apatite deposition, a key process in bone mineralization (Mizukami et al. 2023; Kirsch et al. 2000a). The ability to transport and concentrate calcium and phosphate ions underscores the indispensable role MVs play in regulating bone growth and maintaining bone integrity (Anto et al. 2024; Arivalagan et al. 2023).

The size of MVs is a critical characteristic that influences their biological functions, particularly their interactions with cells and tissues. The diameter of MVs typically ranges from 50 to 200 nm (Welsh et al. 2024; Vaiaki and Falasca 2024), with 8% of studies reporting particles smaller than 200 nm (17). Size distribution is crucial for the role MVs play as carriers of mineralizing components given that smaller MVs may penetrate tissue barriers and interact with cells more effectively (Vaiaki and Falasca 2024).

An underexplored and debatable aspect of research into MVs is genetic content. While only a few studies have investigated the RNA load of MVs, understanding their mRNA and miRNA profiles could provide valuable insights into the regulatory mechanisms underlying their functions (Mizukami et al. 2023; Sawada et al. 2007; Arivalagan et al. 2023; Li et al. 2016; Chaturvedi et al. 2015). Two studies reported the presence of RNA in the composition of MVs (Arivalagan et al. 2023; Chaturvedi et al. 2015), which may play roles in cellular processes such as differentiation, apoptosis, immune response, and pathological calcification.

MVs hold potential applications in regenerative medicine, particularly in gene and drug delivery, and as diagnostic tools. However, their role in initiating pathological calcification has raised concerns (Roszkowska et al. 2018; Chaturvedi et al. 2015; Chen et al. 2018, 2021). For example, MVs contribute to VSMC calcification by concentrating calcium and phosphate ions, promoting crystal formation mediated by enzymes like ALP (Chaturvedi et al. 2015; Chen et al. 2021, 2010a). This duality highlights the need for further research into harnessing MVs for therapeutic purposes while mitigating their potential adverse effects in pathological conditions. A study on MVs isolated from VSMC revealed that the RhoA/Rho kinase (ROCK) signaling pathway influences the uptake of calcification inhibitors such as fetuin-A, directly impacting the progression of vascular calcification (Chen et al. 2018). This modulation of fetuin-A by ROCK underscores the dynamic nature of MVs, positioning them as active modulators rather than passive carriers in calcification processes (Chen et al. 2010a).

Additionally, MVs can retain reactive oxygen species (ROS), which significantly influence their ability to induce calcification (Chaturvedi et al. 2015; Chen et al. 2021). Oxidative stress, driven by ROS, exacerbates calcification by destabilizing the membrane of MVs and promoting mineral deposition through oxidative pathways. These findings highlight that ROS is a promising therapeutic target to mitigate pathological calcification (Chen et al. 2021).

Interestingly, verapamil, an L-type inhibitor of calcium channels, suppresses both calcification and the activity of MVs in bovine VSMC (Chen et al. 2010b). Proteins such as collagen (Roszkowska et al. 2018) and annexins (Chen et al. 2008) also play crucial roles in mineralization mediated by MVs in VSMC. Moreover, studies have indicated that stabilizing cellular microtubules can inhibit the release of MVs and osteogenic signaling, suggesting that targeting the stability of MVs and associated signaling pathways could offer viable therapeutic strategies to combat vascular calcification​ (Lee et al. 2014).

Finally, this review underscores the critical need for standardizing protocols in research into MVs. Establishing unified guidelines for the isolation, characterization, and functional analysis of MVs would greatly enhance the reproducibility of results and accelerate the translation of therapies based on MVs into clinical applications. Additionally, given that most of the current studies rely on animal models, expanding research to human sources is imperative to deepen our understanding of the translational relevance of MVs in human health and disease.

The digital library developed as part of this review was designed to facilitate collaboration and innovation in MV research. This platform offers a feasible way for researchers to access, filter, and explore data collected from a high number of studies. To further enhance its utility and promote collaborative participation, we propose the following strategies:

1. Initial feedback from early users of the digital library has been overwhelmingly positive, particularly regarding its user-friendly interface and the ability to filter data efficiently. For example, one research team highlighted how the platform streamlined their comparative analysis of MV isolation protocols across tissue types, saving valuable time.

2. Feedback and submission portal: Researchers will be able to share data, suggest new metadata categories, and provide insights for improving the library.

3. Institutional partnerships: Collaboration with academic institutions and research consortia will populate the database with high-quality data.

4. Training initiatives: Workshops and webinars will introduce the library to researchers, demonstrating its functionalities and encouraging its adoption.

These measures aim to establish the digital library as a cornerstone resource for MV research, enabling cross-disciplinary collaboration and driving innovation.

Conclusion

This systematic review provides a comprehensive overview of MV research, highlighting the critical need for standardized protocols to isolate and characterize MVs. Such standardization is essential to improve reproducibility and comparability across studies. The findings confirm that digestion with collagenase combined with centrifugation remains the most effective isolation method. Additionally, ALP activity assays and TEM are reliable characterization techniques, while complementary methods such as NTA and proteomics offer further insights into MV functionality and composition.

MVs play pivotal roles in physiological and pathological mineralization. For instance, in physiological contexts, MVs facilitate bone mineralization by serving as nucleation sites for hydroxyapatite formation. In pathological conditions, they contribute to vascular calcification, as seen in atherosclerosis, where MVs promote the deposition of calcium phosphate minerals in vascular tissues. These dual roles underscore their potential as both therapeutic tools and targets in regenerative medicine and disease management. Their enrichment with proteins such as ALP, annexins, and osteocalcin underscores their utility in bone regeneration and tissue engineering. However, the translational potential of MVs in regenerative medicine and disease management necessitates further research, particularly on human-derived MVs.

The establishment of an open-access digital library consolidating MV data represents a step forward in the field. By enabling comparisons across isolation protocols, tissues, and cell types, this platform provides a quick search facility with a robust foundation for future investigations into MV-mediated mineralization. Researchers are encouraged to contribute to and utilize this resource to foster collaboration and innovation, ultimately driving progress in therapeutic applications and clinical translation.

Author contribution

All authors contributed to the study conception and design. The idea of the study was conceived by Pietro Ciancaglini, José Eduardo Santarem Segundo, and Ana Paula Ramos. Material preparation and data collection analysis were performed by Gildacio Pereira Chaves Filho, Pedro Tavares, Ananda Fernanda de Jesus, José Eduardo Santarem Segundo, and Ana Paula Ramos. The first draft of the manuscript was written by Gildacio Chaves Filho, and all the authors commented on previous versions of the manuscript. All the authors have read and approved the final manuscript.

Funding

This study was supported by the São Paulo State Research Foundation (FAPESP; grants 2019/25054–2 and 2019/08568–2 and scholarships 2021/03349–0 and 2023/07244–4). The authors thank the Provost Office of Undergraduate Studies of the University of São Paulo for the scholarship granted to Pedro Tavares. This study was also supported by the National Council for Scientific and Technological Development- Brazil (CNPq grant 408540/2021–1). The authors thank Cynthia Manso for reviewing the English language. P.C., J.E.S.S., and A.P.R. are CNPq researchers.

Data availability

The datasets generated during and/or analyzed during the current study are available in the Mineralizing Vesicles Digital Library repository, http://200.144.245.117/mineralizing-extracellular-vesicles.

Declarations

Competing interests

The authors declare no competing interests.