If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
National Children's Research Centre, Children's Health Ireland, Dublin 12, IrelandSchool of Pharmacy and Biomolecular Sciences, RCSI University of Medicine and Health Sciences, Dublin 2, Ireland
National Children's Research Centre, Children's Health Ireland, Dublin 12, IrelandSchool of Pharmacy and Biomolecular Sciences, RCSI University of Medicine and Health Sciences, Dublin 2, Ireland
National Children's Research Centre, Children's Health Ireland, Dublin 12, IrelandDepartment of Paediatrics, RCSI University of Medicine and Health Sciences, Dublin 2, Ireland
Corresponding author at: School of Pharmacy and Biomolecular Sciences, Royal College of Surgeons in Ireland, 123St. Stephen's Green, Dublin 2, Ireland.
National Children's Research Centre, Children's Health Ireland, Dublin 12, IrelandSchool of Pharmacy and Biomolecular Sciences, RCSI University of Medicine and Health Sciences, Dublin 2, Ireland
We performed an initial characterisation of EVs in serum from PWCF demonstrating the potential of serum EVs for further diagnostic investigation.
•
This study to our knowledge is the first to analyse serum EVs in CF using both NTA and proteomics.
•
EVs were successfully isolated in greater numbers with size exclusion chromatography compared to density ultracentrifugation from PWCF using low blood volumes (250 ul).
•
The study has revealed interesting changes in extracellular cargo of EVs as CF disease progresses and when PWCF were treated with Kaftrio establishing a clinical potential for serum EVs in CF.
Abstract
Background
Extracellular vesicles (EVs) are emerging as biomarkers of disease with diagnostic potential in CF. With the advent of highly effective modulator therapy, sputum production is less common and there is a need to identify novel markers of CF disease progression, exacerbation and response to therapies in accessible fluids such as serum.
Methods
We used size exclusion chromatography (SEC) to isolate and characterise EVs from the blood of PWCF of different ages and compared to ultracentrifugation (UC). We used nanoparticle tracking analysis to measure the number of EVs present in serum obtained from children and adults with CF. Mass spectrometry based proteomics was used to characterise protein expression changes between the groups.
Results
EVs were successfully isolated in SEC fractions from 250 µl serum from PWCF in greater numbers (p <0.01) than density ultracentrifugation. There was not a significant difference in EV numbers between young children with CF and controls. However, there was significantly more EVs in adults compared to children (<6yrs) (p < 0.05). EVs from PWCF before and after Kaftrio treatment were also analysed. Significant protein expression changes were observed within all 3 group. The largest changes detected were between children and adults with CF (57 proteins had a 1.5 fold change in expression with 19 significant changes p < 0.05) and PWCF taking Kaftrio (24 significant changes in EV protein expression was observed 12 months post treatment).
Conclusion
In this pilot study, we performed an initial characterisation of EVs in serum from PWCF demonstrating the potential of serum EVs for further diagnostic investigation.
Extracellular vesicles (EVs) are small vesicles released from most living cells enriched with protein, RNA and lipid cargo that enables communication with other cells. The ability of EVs to bear disease-specific and often inflammatory signatures is increasingly recognized across several diseases [
]. We and others discovered that EVs are produced in large amounts by Cystic Fibrosis (CF) airway cells, can drive neutrophil migration into the airways, and that treatment with CFTR modulators can impinge on EVs release [
]. In our previous study, we isolated EVs from bronchoalveolar lavage fluid (BALF) with good recovery in children under a routine surveillance program [
] so there is a need to identify novel markers of CF disease progression, exacerbations and response to therapies in accessible fluids. Serum is a good source of EVs which are regularly released into the bloodstream upon cellular activation and potentially represent an indicator of systemic disease in persons with CF (PWCF) [
EVs can be isolated by different methods that exploit their physical or chemical properties including ultracentrifugation, size exclusion, affinity exclusion, precipitation and recently microfluidics. All techniques have advantages and shortcomings and it remains a challenge within the EV field to obtain well-defined EV fractions of high quantity and purity with short isolation times [
]. Size exclusion chromatography (SEC) is a chromatographic method which allows good separation of large molecules from small molecules with minimal volumes. We hypothesized this may be a good method to isolate EVs from the blood of children where low volumes of blood are drawn. In this small study, we examined the potential of size exclusion chromatography to isolate and characterise EVs from the blood of PWCF of different ages and compared to ultracentrifugation (UC) a commonly used EV isolation technique.
2. Methods
More detailed methods are available in Supplementary Materials.
2.1 Subjects
Blood and BALF was obtained through The Study of Host Immunity and Early Lung Disease in Cystic Fibrosis (SHIELD-CF) or RECOVER (Real World Clinical Outcomes with Novel Modulator Therapy Combinations in People with CF) from children with CF at Children's Health Ireland at Crumlin (CHI) and from adults with CF attending St. Vincent's University Hospital (SVUH). All study participants were recruited as approved by the Ethics Medical Research Committee at CHI or SVUH.
2.2 EV fraction isolation from serum by SEC
Briefly, EVs were processed using an AKTA Purifier (GE Healthcare), 250 µl of serum was applied to a Capto Core 700 HiTrap column equilibrated with 5 mL of 1X PBS and fractions eluted with 1.0 mL of 1X PBS before concentrating fractions with 100 kDa filters.
2.3 EV isolation by ultracentrifugation
Extracellular vesicle (EV) fractions were isolated by differential ultracentrifugation. Briefly, serum (250 μl) was centrifuged at 120,000 g for 2 h 4 °C (XL-70 ultracentrifuge, Beckman-Coulter, Villpinte, France) to pellet the EVs.
2.4 Protein extraction and immunoblotting
SEC and UC EV fractions were analysed for EV markers, CD9, flotillin-1 (Flot-1), as well as Apolioprotein B1(Apo-B1) (Cell Signalling) as previously described [
]. EV fractions were analysed for further EV markers using an antibody array called Exoarray (Systems Biosciences) as per manufacturer's instructions.
2.5 Nanoparticle tracking analysis (NTA)
Particle size distribution in cellular supernatants and SEC and UC samples was determined by NTA using a NanoSight NS300 system (Malvern Technologies) as previously described [
The Q-exactive raw data files were de novo sequenced and cross searched against a Human UniProtKB database Proteins were quantified by Label-free Quantification using MaxQuant software as previously described [
Analysis was conducted using Prism 9 (GraphPad Prism, San Diego, CA, USA). A non-parametric t-test (Mann Whitney) was used for analysing two different variables to allow for small samples number without the assumption of normal distribution. A Kruskal-Wallis one-way ANOVA was used for statistical analysis involving three or more groups.
3. Results
Serum was obtained from a donor without CF and analysed by SEC to optimise the method. Three peaks were detected in the UV trace (Supplementary Fig. 1) and their respective fractions collected. Peak A (1–9), Peak B (1–6) and Peak C (1–5) fractions were analysed by NTA. The largest particle number/ml was observed in Peak A where 65.1% of total particles were recovered compared to 30.1% in Peak B and 4.8% in Peak C (Fig. 1b). The largest particle sizes were also observed in Peak A fractions (Fig. 1c). An increase in EV particles were observed in both A and B fraction compared to UC which was significant when A&B particle numbers were combined (p < 0.05) (Supplementary Fig. 2) (Fig. 1d). Transmission Electron Microscopy (TEM) was used to confirm the size of the vesicles in the pooled peak fractions. EVs were mostly detected between 50 and 250 nm with minimal EVs observed in Peak C (Fig. 1e). The expression of exosomal markers and lipoprotein contaminant Apo-B1 were analysed by immunoblotting. CD9 and Flotillin were strongly present in the fractions (Peak A&B). However, there was some Apo-B1 present in these fractions compared to UC EVs (Fig. 1f) (Supplementary Fig. 3). Less CD9 but more Apo-B1 was detected in the Peak C1/C2 fractions. Based on this finding, combined with observations from the TEM/NTA data on low EV number in Peak C, we excluded it for clinical studies. Further validation of EV markers in the pooled fractions from SEC and UC was performed using exosome antibody array. ALIX, TSG101, ICAM, ANAX5 as well as CD81, and CD63 were detected in the SEC and UC fractions with higher GM130 in UC fractions (Fig. 1f).
Fig. 1Serum EVs isolated by SEC were characterised by NTA, TEM and immunoblotting
A. Schematic of workflow from sample isolation to EV characterisation B. Peak A(1–9), Peak B(1–6) and Peak C (1–5) serum fractions were analysed by Nanoparticle Tracking Analysis (NTA). Fractions were quantified in triplicate with particle number/ml displayed for individual peaks. C. Size distribution of particles isolated in Peak A-C fractions. D. NTA analysis was performed on A and B pooled fractions (SEC) and compared to UC fraction. Histograms represent the average of three experiments. Statistical analysis used was unpaired student t-test, and error bars denote mean± SD E. Representative TEM images of combined SEC Peak A, Peak B fractions and UC fractions at 30,000 60,000 and 200,000 magnification. F. Protein expression of CD-9, Flotillin-1 and Apolipoprotein B-1 was determined by immunoblotting in the fractions isolated by SEC G. Protein expression of Flotillin-1, ICAM, ALIX, CD-81, CD63, EPCAM, ANXA5, TSG101& GM103 was determined by and EV antibody array in Peak A and Peak B combined fractions and UC pellet.
Our objective was then to test the SEC methods in serum from PWCF. We obtained 250 µl of serum from paediatric CF and control donors in the 1–4year age group with minimal disease (see Supplementary Table 1 for clinical characteristics). EVs were isolated by UC and SEC (pooled A&B). Although there was variability in the concentration of serum EV particles isolated by both methods from control (Fig. 2a) and CF donors (Fig. 2c) there was a significantly higher number of EVs isolated by SEC compared to UC in both control (p < 0.01) (Fig. 2b) and CF (Fig. 2d) (p < 0.01) samples. There was not a significant difference in the total number of EVs from CF and control donors of this age by either SEC (Fig. 2e) or UC (Fig. 2f). When we compared EVs obtained from BALF with EVs obtained from the serum of CF and controls donors (Fig. 2g) of this age group there was no significant differences observed either.
Fig. 2Serum EV fractions isolated from PWCF by SEC demonstrated an increased number of EVs compared to UC isolated fractions
A. EVs were isolated from 6 control paediatric donors (1–4 yrs) using SEC (combined Peak A&B) and UC and analysed by NTA. A. Individual control donor particle numbers are displayed. B. The average control donor particle number/ml of the 6 donors for each method is displayed by whisker plot. Statistical analysis used was the Mann Whitney test, and error bars denote mean±SD C. Individual CF donor particle numbers are displayed. D. The average CF donor particle number/ml of the 6 donors for each method is displayed by whisker plot. Statistical analysis used was the Mann Whitney test, and error bars denote mean±SD. E. Average particle number for control donors is compared to CF donors (SEC) by whisker plot. F. Average particle number for control donors is compared to CF donors (UC) by whisker plot. Statistical analysis used for D and E was the Mann Whitney test, and error bars denote mean±SD. G. Serum EVs isolated by UC (control and CF donors) were compared to BALF EVs isolated by UC. Statistical analysis used for D and E was a Kruskal Wallies-Dunn test and error bars denote mean±SD.
EVs were isolated from serum of CF donors compared to controls (4–6yrs) and measured by NTA without significant difference observed in this age group (Fig. 3a). However, there was altered EV protein expression observed between the CF and control groups (4–6yrs). 180 proteins were identified in total, 45 proteins changed in expression by 1.5 fold between the groups and 8 were significantly different (p < 0.05) (Fig. 3b) (Supplementary Table 2/Fig. 6a). EVs were also obtained from the serum of adults (18yrs+). Significantly higher EV numbers were seen in adults compared to children (Fig. 3c) (p < 0.05). 171 proteins were identified by MS in which 52 proteins had a 1.5 fold change in expression between the paediatric and adult groups and 19 proteins changed significantly (p < 0.05) (Fig. 3d) (Supplementary Table 2/Fig. 6b). We did not observe a significant change in EV numbers by NTA in PWCF 12 months post Kaftrio treatment, (Fig. 3e) but 62 proteins changed in expression between these 2 groups with 24 significant (Fig. 3f) (Supplementary Fig. 6c). Bioinformatic analysis of all three datasets revealed an enrichment in vesicle/extracellular proteins (Supplementary Fig. 4) with 81% of protein present in the vesiclepedia compendium. FunRich analysis identified immune response to be enriched significantly in all EV protein datasets and haemostasis was a significantly enriched biological pathway. Principal component analysis (PCA) of protein quantification data significantly changing in abundance (p < 0.05) was performed (Supplementary Fig. 5) for all datasets. Some differentiation into separate groups was observed for the adult data sets with less detected in the children versus control group.
Fig. 3Differential protein expression and EV number was observed between paediatric and adult cohorts with CF
A. EVs were isolated by SEC from 5 paediatric CF and control donors and NTA analysis performed. Whisker plot displayed represent average of 5 donors. Statistical analysis used was the Mann Whitney test and error bars denote mean±SD. B. EV fractions from 5 paediatric CF and control donors were analysed by mass spectrometry (proteomics). Heat map of 8 proteins is displayed for CF/Control donors. Analysis is by label free quantification (MaxQuant). C. EVs isolated by SEC from 5 paediatric CF donors were compared to 5 adults with CF. Whisker plot displayed represents average of 5 donors. Statistical analysis used was a Mann Whitney test and error bars denote mean±SD. D. EV fractions from 5 paediatric CF and adult donors were analysed by mass spectrometry (proteomics). A heat map of changing proteins is displayed for CF Paediatric/Adult donors. E. EVs isolated by SEC from 5 adult CF donors before and 12 months after Kaftrio treatment. Whisker plot displayed represents average of 5 donors. Statistical analysis used was Wilcoxin test and error bars denote mean±SD. F. EV fractions from the drug treated groups were analysed by mass spectrometry (proteomics) for all donors. A heat map of changing proteins is displayed for 4 donors (technical duplicate) with an outlier removed.
In this study, we hypothesized that SEC may be a good method to isolate EVs from the blood of PWCF and be able to differentiate between early and more progressive disease. We successfully identified SEC fractions rich in EV/markers from serum. Some advantages of EV isolation by SEC over density UC included increased EV number using small volumes of blood (250 µl) in both control and CF samples (Fig. 2), lower processing time (30 min SEC/fraction concentration compared to 2 hrs minimum UC) and cost effective reusable columns. The identification of apolipoproteins in blood EVs by SEC has been reported by several groups [
Comparison of small extracellular vesicles isolated from plasma by ultracentrifugation or size-exclusion chromatography: yield, purity and functional potential.
Size-exclusion chromatography as a stand-alone methodology identifies novel markers in mass spectrometry analyses of plasma-derived vesicles from healthy individuals.
]. Additionally, the presence of apolipoproteins on the surface of EVs and changes in their abundance could be indicative of specific disease pathologies. Therefore, they could be considered as more than EV contaminants.
Although we did not observe an increase in the numbers of EVs from the serum of young children with CF compared to controls, the expression of inflammatory proteins such as neutrophil activating peptide within CF EVs increased. This echoes previous reports of early neutrophil inflammatory markers detected in young children with CF [
Australian respiratory early surveillance team for cystic fibrosis (AREST CF). infection, inflammation, and lung function decline in infants with cystic fibrosis.
Am J Respir Crit Care Med.2011; 184 (Jul 1): 75-81
]. Significant changes in protein expression during disease CF disease progression and Kaftrio treatment were detected and both mapped to immune response pathways. Interestingly significant enrichment in haemostasis pathways were also observed indicating presence of platelet EVs previously reported by Pavlainen et al. [
]. While predominant platelet or RBC EVs may reduce the identification of lower abundant CF related EV proteins, they may also provide new insights into CF pathophysiology as platelets are dysregulated and play an inflammatory role in CF [
To conclude 1. this is the first study of EVs in serum from PWCF and controls (>30 donors) to our knowledge hopefully leading to future studies in the area. Larger donor number and longitudinal time points should further establish the diagnostic potential of serum EVs in CF. 2. EVs were successfully isolated in greater numbers by SEC compared to UC using low blood volumes more suitable for children and faster processing times. 3. The study revealed interesting changes in extracellular cargo of EVs during disease progression and Kaftrio treatment establishing the clinical potential of serum EVs in CF.
Funding sources
This publication has emanated from research conducted with the financial support of The National Children's Research Centre under Project Grant No C/17/3. Additional support emanated from the Comprehensive Molecular Analytical Platform under The SFI Research Infrastructure Programme, reference 18/RI/5702.
Authorship contributions
Conception and Design: AT, JAC
Experiments: AT, KW, NL
Clinical Data Collection and Support: SC, EMK, PM
Analysis and Interpretation: AT, JAC, ED, NL
Manuscript Preparation: JAC, AT, NL, SC, PM
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Comparison of small extracellular vesicles isolated from plasma by ultracentrifugation or size-exclusion chromatography: yield, purity and functional potential.
Size-exclusion chromatography as a stand-alone methodology identifies novel markers in mass spectrometry analyses of plasma-derived vesicles from healthy individuals.
Australian respiratory early surveillance team for cystic fibrosis (AREST CF). infection, inflammation, and lung function decline in infants with cystic fibrosis.
Am J Respir Crit Care Med.2011; 184 (Jul 1): 75-81
00056-5&id=gr1.jpg){kind=link}
00056-5&id=gr2.jpg){kind=link}
00056-5&id=gr3.jpg){kind=link}