Exosomes are membrane-bound extracellular vesicles involved in intercellular communication and tumor cell metastasis. In this study, flow field-flow fractionation (FlFFF) was utilized to separate urinary exosomes by size, demonstrating a significant difference in exosome sizes between healthy controls and patients with prostate cancer (PCa). Exosome fractions of different sizes were collected for microscopic analysis during an FlFFF run and evaluated with exosome marker proteins using Western blot analysis. The results indicated that exosomes of different sizes originated from different types of cells. Collected exosome fractions were further examined using nanoflow ultrahigh performance liquid chromatography-electrospray ionization-tandem mass spectrometry (nUPLC-ESI-MS/MS) for lipidomic analysis. A total of 162 lipids (from 286 identified) were quantified using a selected reaction monitoring (SRM) method. The overall amount of lipids increased by 1.5- to 2-fold in patients with PCa and degree of increase was more significant in the smaller fractions (diameter <150 nm) than in the larger ones (diameter >150 nm) some classes of lipids. In addition, neutral lipids like diacylglycerol (DAG) and triacylglycerol (TAG) decreased in all exosomes without size dependency. Moreover, a dramatic increase in 22:6/22:6-phosphatidylglycerol (PG) was observed and significant decrease in (16:0,16:0)- and (16:1, 18:1)-DAG species (nearly 5-fold) and high abundant TAG species (>2.5-fold) was observed in patients with PCa. The results of this study indicate that FlFFF can be employed for the high-speed screening of urinary exosome sizes in patients with PCa and lipidomic analysis of the fractionated exosomes has potential for developing and distinguishing biomarkers of PCa.

No abstract

The efficiencies of four different methods for the extraction of phospholipids (PLs) and lysophospholipids (LPLs) from human plasma samples were examined by comparing extraction recovery values using nanoflow liquid chromatography-electrospray ionization-mass spectrometry (nLC-ESI-MS). For recovery measurements, six PL and six LPL standards of different head groups were spiked into a human plasma sample, and the peak areas of each individual species after extraction were measured from the chromatograms of the nLC-ESI-MS runs. Recovery was calculated by comparing the peak area of an extracted standard species with that of the same species' spike after extraction of the same plasma sample. For lipid extraction, four different extraction methods were examined: three based on the Folch method with different organic solvents such as CHCl(3), methyl-tert-butyl ether (MTBE), and MTBE/CH(3)OH, and one relatively fast method involving CH(3)OH only. Evaluations of recovery showed that the modified Folch method with MTBE/CH(3)OH proposed in this study was effective for extracting most PL and LPL standards. Then, the four extraction methods were compared with the identified numbers of plasma PLs and LPLs, of which molecular structures can be confirmed by data-dependent, collision-induced dissociation experiments during nLC-ESI-MS-MS. These results demonstrated that the proposed method yielded the identification of 54 LPLs and 66 PLs from a plasma sample, which was the highest identification rate among the four methods.

In this study, lipid analysis based on isotope-labeled methlylation (ILM) was performed by nanoflow ultrahigh performance liquid chromatography-eletrospray ionization-tandem mass spectrometry (nUPLC-ESI-MS/MS) for enhanced detection and quantification of targeted phospholipids. ILM depends on methylation of phosphate groups by (trimethylsilyl)diazomethane, and the ILM based quantitation with reversed phase nUPLC-ESI-MS/MS provides advantages in PL profiling such as enhanced detectability of methylated PLs owing to increased hydrophobicity and substantial increase in resolution due to the increase of retention. Efficacy of ILM in nUPLC-ESI-MS/MS analysis was evaluated in the selected reaction monitoring (SRM) method by varying the mixing ratio of H-/D-methylated PL standards, which resulted in the successful quantification of 24 species, including phosphatidic acid (PA), phosphatidylserine (PS), phosphatidylglycerol (PG), ceramide-1-phosphate (Cer1P), phosphoinositides, and cardiolipin (CL), with ∼6.6% variation in the calculated ratio of H-/D-methylated PLs. The method was applied to the lipid extracts from a DU145 cell line after D-allose treatment, resulting in the quantification of 83 PLs of which results were not statistically different from those obtained by conventional quantification methods. Morever, detection and quantification of CLs and PAs were evidenced to be highly effective when used with the ILM method as 43 CLs and 20 PAs from cellular lipid extracts were analyzed while only 18 CLs and 12 PAs were identified when conventional methods were carried out. This proves the ILM combined with LC-MS to be a promising method for analysis of the aforementioned classes of lipids. Overall, the study highlighted the applicability of targeted quantification by the ILM method in lipidomic analysis and demonstrated an improvement in the detection of less abundant anionic PLs.

The intertransformation of silver nanoparticles (AgNPs) and ionic silver (Ag(I)) in the environment determines their transport, uptake, and toxicity, demanding methods to simultaneously separate and quantify AgNPs and Ag(I). For the first time, hollow fiber flow field-flow fractionation (HF5) and minicolumn concentration were on-line coupled together with multiple detectors (including UV-vis spectrometry, dynamic light scattering, and inductively coupled plasma mass spectrometry) for full spectrum separation, characterization, and quantification of various Ag(I) species (i.e., free Ag(I), weak and strong Ag(I) complexes) and differently sized AgNPs. While HF5 was employed for filtration and fractionation of AgNPs (>2 nm), the minicolumn packed with Amberlite IR120 resin functioned to trap free Ag(I) or weak Ag(I) complexes coming from the radial flow of HF5 together with the strong Ag(I) complexes and tiny AgNPs (<2 nm), which were further discriminated in a second run of focusing by oxidizing >90% of tiny AgNPs to free Ag(I) and trapped in the minicolumn. The excellent performance was verified by the good agreement of the characterization results of AgNPs determined by this method with that by transmission electron microscopy, and the satisfactory recoveries (70.7-108%) for seven Ag species, including Ag(I), the adduct of Ag(I) and cysteine, and five AgNPs with nominal diameters of 1.4 nm, 10 nm, 20 nm, 40 nm, and 60 nm in surface water samples.