Comparative characterisation of extracellular vesicles from canine and human plasma: a necessary step in biomarker discovery

Extracellular Vesicles (EV) have become an interesting focus as novel biomarkers of disease and are increasingly reported upon in humans and other species. The Minimal Information for Studies of Extracellular Vesicles 2018 (MISEV2018) guidelines were published to improve rigor and standardisation within the EV field and provide a framework for the reliable isolation and characterisation of EV populations. However, this rigor and standardisation has been challenging in the area of comparative medicine. Herein we present the successful isolation of EVs from human and canine plasma using Size Exclusion Chromatography and characterise these EVs according to best international practice. This study provides evidence for the reliable comparison of human and canine EVs isolated by this approach, and a baseline description of the EVs from healthy dogs to inform future biomarker studies. This work also demonstrates that the MISEV2018 guidelines can be successfully applied to EVs isolated from canine plasma. Supplementary Information The online version contains supplementary material available at 10.1007/s11259-024-10405-0.


Introduction
Circulating small Extracellular Vesicles (sEVs), measuring 30-150 nm, are potential biomarkers of disease, as their composition and cargo reflect their cell of origin (Yuana et al. the recommendation that 'the need to demonstrate presence of EV markers and absence or depletion of putative contaminants… can be generalised to all species, cells and conditions' (Thery et al. 2018), and the wide support of the MISEV2018 guidelines amongst the EV community (Witwer et al. 2021).
In this report, we present the successful isolation and characterisation of plasma derived EVs from canine and human plasma using SEC, according to the MISEV2018 guidelines.This work provides a framework for other canine EV based comparative studies.

Recruitment and plasma preparation
Canine plasma samples were obtained from 10 healthy dogs undergoing routine veterinary care, once all clinical diagnostics were complete, in accordance with the protocol approved by the Animal Research Ethics Committee, University College Dublin (UCD; Ref: AREC-E-20-26-Kelly).No samples from dogs were taken specifically for this study, and all dogs remained in the care of their owners throughout.The samples were pooled to enable the replication of experiments across technical replicates, given the constraint of very small volumes remaining after clinical diagnostics.The pooled plasma was then aliquoted and stored at -80 °C.
Healthy human controls (Supplementary Table 1) were recruited from patients attending the outpatient Department at the Mater Misericordiae University Hospital, Dublin, Ireland, following Institutional Review Board ethical approval (Ref: 1/378/2189).Full informed consent was obtained, and a EDTA plasma sample was obtained and spun at 3000xg for 10 min to separate plasma, and subsequently at 2500xg to remove platelets.Platelet poor plasma was then aliquoted and stored at -80 °C for subsequent Extracellular Vesicle (EV) isolation.

Extracellular vesicle (EV) isolation
EVs were isolated from plasma, using Size Exclusion Chromatography (SEC), using iZon qEVoriginal Gen2 70 nm Columns (iZon Science).Isolations from both human and canine plasma were performed using an Automated Fraction Collector (iZon Science).For each isolation, the column was flushed with freshly filtered PBS, and 500µL plasma was added to the loading frit.A buffer volume of 100µL was allowed to pass before collection of 13 fractions each of 400µL elute were collected for downstream analysis.

Nanoparticle tracking analysis (NTA)
Nanoparticle Tracking Analysis (NTA) was performed using a NanoSight NS300, to determine EV size and concentration from the individual EV fractions.The NTA instrumentation was configured with a 488 nm laser and a high sensitivity CMOS camera.Samples were diluted in freshly filtered PBS (1:25 dilution) and analysed under constant flow conditions (flow rate = 50) at 25 °C, camera level 10-12 and screen gain 5. Five x 60 s videos were captured for each sample, and human and canine EVs were analysed in triplicate.Data were analysed using NTA 4.1 software, with a detection threshold of 10 and bin size of 2.

EV lysis and protein quantification
EVs were lysed prior to establishing the protein concentration in each SEC fraction.A volume of 50µL elute from each fraction were collected and lysed with 10µL of a Triton-based lysis buffer [50mM Tris HCL pH 7.4, 150mM NaCl, 1% Triton, 1% EDTA, 1mM phenylmethylsulphonyl, 1% phosphatase inhibitor, 1% protease inhibitor cocktail].Samples were incubated on ice for 30 min, vortexing every 10 min, followed by water bath sonication for 3 min and centrifugation at 10,000xg for 20 min.Protein quantification was then performed using the Pierce BCA Protein Assay (ThermoFisher,#23,227) kit in duplicate according to manufacturer's instructions, with concentration determined from a Bovine Serum Albumin (BSA) standard curve.

Transmission electron microscopy
Once the optimal purified collection volume (PCV) of the plasma EVs had been determined, four EV fractions were pooled for further analysis of morphology using TEM.From the PCV, 400µL of EVs were then concentrated using a 10 kDa cut-off centrifugal filter (Amicon, #UFC501096).10µL of EVs were placed on a formvar carbon-coated copper EM grid for 60 min.Vesicle coated grids were washed three times with PBS and then fixed using 2.5% glutaraldehyde for 10 min.The grids were then washed in distilled water, before staining with 2% uranyl acetate for 15 min, and embedded in methyl cellulose-UA for 10 min on ice.Excess cellulose was removed, and the grids allowed to air dry.TEM was then performed using a FEI Tecnai 120 microscope, operating at an accelerating voltage of 120 kV.Images were taken of the entire field at 87000X.

Western blot analysis for extracellular vesicle markers
EV isolations were performed as described in triplicate from the pool of canine plasma and from three individual human controls, collecting the optimal PCV.These were lysed and protein quantified as described above.A concentrate of 30 µg of protein was combined with 5µL 1xSDS loading buffer (New England BioLabs) and 1.25 M DTT and heated at 95 °C for 5 min.Proteins were run on an 8-12% SDS NuPAGE Bis-Tris gel (Invitrogen, #NP0321BOX) in a NuPAGE MOPS SDS Running buffer (Invitrogen, #NP0001) at 200 V for 42 min.Resolved proteins were then transferred to a nitrocellulose membrane at a constant 50 V for 80 min.The effective transfer was confirmed using a Ponceau stain, and membranes were then blocked in 1x TBS containing 5% (w/v) BSA.Proteins were detected by incubation with primary antibodies; Alix (abcam, ab186728, 1/1000), Calnexin (abcam, ab 112,995, 1/2000), HSP70 (abcam, ab181606, 1/1000), CD63 (abcam, ab271286, 1/1000), ApoA1 (abcam, ab211472, 1/100) in blocking solution overnight at 4 °C.Membranes were then washed in TBS with 0.1% Tween, and incubated in the appropriate species IRDye-conjugated secondary antibody (Li-COR Biosciences, IRDye-680RD conjugated goat anti-rabbit IgG #925-68071 and IRDye-800CW goat anti-mouse IgG #925-32210) for 1 h in the dark at room temperature.Proteins were then visualised by scanning the membrane with an Odyssey CLX Infrared Imaging System (Li-COR Biosciences) and processed using Image Studio Lite (Li-COR Biosciences, v5.2.5).

Determination of the optimal purified collection volume to allow comparison of canine and human EVs
Nanoparticle Tracking Analysis (NTA) and protein quantification using the Bicinchoninic Acid (BCA) protein assay allowed optimisation of the collection schedule from the iZon Gen2 qEVoriginal Size Exclusion Chromatography (SeC) column to enrich for small Extracellular Vesicles (sEVs) from canine and human plasma (Fig. 1).Peak concentration of particles was seen in the elute collected in the Purified Collection Volume (PCV) consisting of the 4 × 400µL fractions collected following a buffer volume of 2500µL.Protein concentration in the subsequent fractions isolated from both species increased, indicating co-isolation of protein contaminants.This is further supported upon examination of the purity of individual fractions as seen in Fig. 1c and d, using a ratio of the number of particles as measured on NTA per µg of protein, in which a higher ratio is suggestive of a higher degree of purity (Webber and Clayton 2013).

Comparison of canine and human EV morphology by TEM
Size distribution profiles of the PCV fractions of both species demonstrated that the elute is predominantly composed of particles < 200 nm in size, with particles > 200 nm contributing a smaller concentration, indicating successful enrichment for sEVs (Fig. 2).The PCVs isolated from dogs and humans consisted of particles with a mean modal size of 127.5 ± 21.6 nm and 165.0 ± 7.8 nm respectively (p = 0.10, Mann-Whitney test).Transmission Electron Microscopy demonstrates the presence of particles with a typical cupshaped morphology and size consistent with EVs from both species.Small (< 200 nm) and larger (> 200 nm) EVs can be seen (black arrows), as well as lipoproteins (red arrows).

Characterisation of canine and human EVs using recognised EV markers
Extracellular Vesicles were isolated from the pooled plasma collected from n = 10 healthy dogs and from individual samples from three healthy humans.These EVs were then concentrated using centrifugal filtration, and presence and absence of relevant EV markers were analysed by Western Blot (Fig. 3).The presence of CD63, a tetraspanin and non-tissue specific transmembrane protein, and HSP70, a cytosolic protein promiscuously incorporated in EVs, are both clearly demonstrated in the EV preparations from both species.The absence of calnexin, a protein associated with the Endoplasmic Reticulum and Golgi apparatus and not enriched in the small EV populations originating from Multivesicular Bodies, supports the assertion that the PCV is enriched for small EVs < 200 nm.However, the presence of non-EV co-isolated structures and contaminants is also apparent in both species, as demonstrated by the presence of ApoA1.biological samples obtainable from some species (Lawson et al. 2017;Howard et al. 2022) have made the direct comparison of canine and human EVs challenging to date.However, when attempting to address these challenges, it is important to also consider species-specific factors.For instance, the differential bands observed in the Western blot of CD63 between humans and dogs (Fig. 3) suggest the influence of additional variables, such as differential glycosylation (Tominaga et al. 2014).
Lipoproteins were co-isolated with the EVs from both species when using SEC, as seen in the TEM (Fig. 2) and Western Blot (Fig. 3) images presented.Successful removal of lipoprotein contaminants from plasma EVs remains a challenge for researchers in the field (Thery et al. 2018, Takov et al. 2019).Despite this, SEC is commonly used in plasma-derived EV isolation (Royo et al. 2020) as it is reported to minimally alter the EV preparation preserving the functionality of EVs (Mol et al. 2017), and co-isolate

Discussion
Our results show the successful isolation and characterisation of Extracellular Vesicles (EVs) from canine and human plasma, in accordance with the Minimal Information for Studies of Extracellular Vesicle 2018 (MISEV2018) guidelines.Similarly, this also demonstrates that EVs isolated from both species using Size Exclusion Chromatography (SEC) can be reliably compared, as they demonstrate similar elution and size profiles, and can be profiled using accepted characteristic EV markers.Dogs are a popular species for comparative medicine and oncology studies (Paoloni and Khanna 2007).As EV formation and release appears to be evolutionarily conserved between species (Lawson et al. 2017), EVs have become an attractive focus for novel biomarkers of disease in comparative studies.However, the lack of availability of species-specific antibodies and the small volume of available no exception.However, non-EV co-isolated structures can be falsely detected as EVs in many common quantification methods such as NTA (Brennan et al. 2020) and may alter results of functional studies (Takov et al. 2019).As a result, it is essential to include an assessment of these contaminants to inform the purity and reliability of a preparation before any conclusions about the actions or influence of the EVs may be drawn.

Conclusions
In conclusion, we have fully characterised the Extracellular Vesicles (EVs) isolated from the plasma of healthy dogs and human donors, according to the Minimal Information for Studies of Extracellular Vesicle 2018 (MISEV2018) less other soluble factors (Wallis et al. 2021).In addition, it is inexpensive, quick and requires minimal additional equipment, making it a strong candidate for biomarker discovery and eventual clinical translation (Veerman et al. 2021).
Nonetheless, the presence of these contaminants highlights the necessity of EV characterisation when preforming EV population studies.Whilst many groups have successfully isolated canine EVs using various isolation methods, few have fully characterised them, with many studies reporting features such as positive markers and concentration, without an assessment of preparation purity (Kuwahara et al. 2021;da Silva Nunes et al. 2022;Luu et al. 2022).It has been demonstrated that this assessment of non-EV co-isolated components is the least reported upon in the EV literature in general (Poupardin et al. 2021), with the veterinary and comparative biology disciplines being

Fig. 1
Fig. 1 Comparison of the elution profile of canine and human EVs isolated using SEC.Nanoparticle Tracking Analysis (NTA) and BCA protein assay were used to compare the elution profile of EVs isolated from 500µL plasma from the pooled plasma of n = 10 dogs and a single healthy human, each across n = 3 technical replicates.NTA demonstrates the concentration of EVs in each fraction isolated using the qEV Gen2 70 nm Izon column in (A) canine and (B) human plasma,

Fig. 2
Fig.2Size Exclusion Chromatography successfully isolates and enriches for small EVs (<200nm) from canine and human plasma.The size distribution curves of the numbered fractions from the purified collection volume as measured using NTA are shown from (A) canine and (B) human plasma.Both species demonstrate a similar size distribution profile across the selected fractions, though a higher volume of small EVs (< 200 nm) was eluted in Fraction 2, after a buffer volume of 2500µL in the canine EVs (A) whereas the majority of small EVs (200 nm) were eluted in Fraction 3 and 4 for the human sample

Fig. 3
Fig. 3 Comparative Western Blot analysis of isolated canine and human EVs showing similar EV marker profile.EVs isolated from pooled plasma of dogs (n = 10) in triplicate and individual human (n = 3) donors were lysed, concentrated, and analysed using western