Isolation and Characterization of Extracellular Vesicles in Human Bowel Lavage Fluid

Colorectal cancer (CRC) is the third most common cancer worldwide and is detected in late stages because of a lack of early and specific biomarkers. Tumors can release extracellular vesicles (EVs), which participate in different functions, such as carrying nucleic acids to target cells; promoting angiogenesis, invasion, and metastasis; and preparing an adequate tumor microenvironment. Finally, bowel lavage fluid (BLF) is a rarely used sample that is obtained during colonoscopy. It presents low variability and protein degradation, is easy to handle, and is representative of EVs from tumor cells due to proximity of the sample collection. This sample has potential as a research tool and possible biomarker source for CRC prognosis and monitoring. In this study, EVs were isolated from human BLF by ultracentrifugation, then characterized by transmission electron microscopy and atomic force microscopy. EV concentration was determined by nanoparticle tracking analysis, and tetraspanins were determined by Western blot, confirming correct EV isolation. RNA, DNA, and proteins were isolated from these EVs; RNA was used in real-time PCR, and proteins were used in an immunoblotting analysis, indicating that EV cargo is optimal for use and study. These results indicate that EVs from BLF can be a useful tool for CRC study and could be a source of biomarkers for the diagnosis and monitoring of CRC.


Introduction
Colorectal cancer (CRC) is the third most common cancer worldwide and the second leading cause of cancer death [1]. It is known that an early diagnosis of CRC improves patient survival, decreasing mortality; however, at present, CRC is usually detected in late stages because of the lack of early biomarkers and detection techniques [2]. Even so, there are different diagnostic techniques that are used in the clinical setting, the most commonly used of which are diagnostic imaging techniques, such as colonoscopies, positron emission tomography, computed tomography, and nuclear magnetic resonance; however, all these techniques are invasive for the patients [3,4]. There are some less invasive techniques; however, these are non-specific for CRC because of their presence in other pathologies, in addition to their presence in late stages of CRC, such as carcinoembryonic antigen (CEA) detection in plasma, as well as the detection of carbohydrate antigen 19-9 (CA19-9), tumor-associated glycoprotein 72 (TAG-72), and tissue polypeptide-specific antigen (TPS) in serum [4]. Finally, stool is also used for CRC diagnosis; the most used diagnosis techniques are fecal occult blood testing, which presents high sensitivity and specificity but also a Table 1. Extracellular vesicle concentration (particles/mL) in healthy, low-risk, high-risk, and cancer samples determined by nanoparticle tracking analysis. Values are represented as mean ± SD. NS: non-significant differences (p 0.690), (ANOVA; p < 0.05). n = 48; for each sample type n = 12 divided into 3 pools of 4 samples each.

RNA Can Be Isolated from Extracellular Vesicles from Bowel Lavage Fluid and Can for Real-Time PCR
RNA can be isolated from EVs from BLF using TRI reagent LS (Table 2). samples presented an RNA concentration of 17.2-3311 ng/mL, low-risk samples pr a concentration of 46.7-460 ng/mL, high-risk samples presented a concentration 2343 ng/mL, and cancer samples presented a concentration of 39.5-8242 ng/mL. sults show non-significant differences between groups.

RNA Can Be Isolated from Extracellular Vesicles from Bowel Lavage Fl for Real-Time PCR
RNA can be isolated from EVs from BLF using TRI reagent LS samples presented an RNA concentration of 17.2-3311 ng/mL, low-risk a concentration of 46.7-460 ng/mL, high-risk samples presented a co 2343 ng/mL, and cancer samples presented a concentration of 39.5-8 sults show non-significant differences between groups. Table 2. Concentration of extracellular vesicle RNA (ng/mL BLF), extracellula BLF), extracellular vesicle protein (µ g/mL BLF), and direct protein from bow protein; µ g/mL BLF) in healthy, low-risk, high-risk, and cancer samples. n ≥

RNA Can Be Isolated from Extracellular Vesicles from Bowel Lavage Fluid and Can Be Used for Real-Time PCR
RNA can be isolated from EVs from BLF using TRI reagent LS (Table 2). Healthy samples presented an RNA concentration of 17.2-3311 ng/mL, low-risk samples presented a concentration of 46.7-460 ng/mL, high-risk samples presented a concentration of 9.5-2343 ng/mL, and cancer samples presented a concentration of 39.5-8242 ng/mL. The results show non-significant differences between groups. Table 2. Concentration of extracellular vesicle RNA (ng/mL BLF), extracellular vesicle DNA (ng/mL BLF), extracellular vesicle protein (µg/mL BLF), and direct protein from bowel lavage fluid (BLF protein; µg/mL BLF) in healthy, low-risk, high-risk, and cancer samples. n ≥ 17; values are represented as mean ± SD. S: significant differences (p 0.000 for EV DNA; p 0.003 for BLF protein); NS: non-significant differences (p 0.431 for EV RNA; p 0.149 for EV protein), (Kruskal-Wallis; p < 0.05).

Healthy
Low This RNA can be used for gene expression determination by real-time PCR. The study of B2M expression levels (Table 3) presented non-significant differences between groups.

DNA Can Be Isolated from Extracellular Vesicles from Bowel Lavage Fluid
DNA can be isolated from EVs from BLF using TRI reagent LS ( Table 2). Healthy samples presented a DNA concentration of 40.3-3213 ng/mL, low-risk samples presented a concentration of 332-2285 ng/mL, high-risk samples presented a concentration of 13.1-1852 ng/mL, and cancer samples presented a concentration of 67.6-2332 ng/mL. The results show significant differences between groups, with a lower DNA concentration in cancer than the other samples.

Western Blot Analyses of Proteins Isolated from Extracellular Vesicles from Bowel Lavage Fluid and Directly from Bowel Lavage Fluid
Proteins can be isolated from EVs from BLF using TRI reagent LS (Table 2). Healthy samples presented a protein concentration of 1.75-74.4 µg/mL, low-risk samples presented a protein concentration of 5.10-31.3 µg/mL, high-risk samples presented a protein concentration of 5.33-58.8 µg/mL, and cancer samples presented a protein concentration of 1.60-41.6 µg/mL. The results show non-significant differences between groups. In addition, proteins can be directly determined from BLF (Table 2). Healthy samples presented a protein concentration of 84.0-1184 µg/mL, low-risk samples presented a protein concentration of 34.9-1302 µg/mL, high-risk samples presented a protein concentration of 54.1-1395 µg/mL, and cancer samples presented a protein concentration of 19.0-9994 µg/mL. The results show significant differences between groups, with a higher BLF protein concentration in cancer samples than the other samples. Finally, these proteins can be studied by Western blot ( Figure 5).

DNA Can Be Isolated from Extracellular Vesicles from Bowel Lavage Fl
DNA can be isolated from EVs from BLF using TRI reagent LS samples presented a DNA concentration of 40.3-3213 ng/mL, low-risk a concentration of 332-2285 ng/mL, high-risk samples presented a con 1852 ng/mL, and cancer samples presented a concentration of 67.6-2 sults show significant differences between groups, with a lower DN cancer than the other samples.

Western Blot Analyses of Proteins Isolated from Extracellular Vesicles f Fluid and Directly from Bowel Lavage Fluid
Proteins can be isolated from EVs from BLF using TRI reagent LS samples presented a protein concentration of 1.75-74.4 µ g/mL, low sented a protein concentration of 5.10-31.3 µ g/mL, high-risk samples concentration of 5.33-58.8 µ g/mL, and cancer samples presented a pr of 1.60-41.6 µ g/mL. The results show non-significant differences betw dition, proteins can be directly determined from BLF (Table 2). Healthy a protein concentration of 84.0-1184 µ g/mL, low-risk samples presente tration of 34.9-1302 µ g/mL, high-risk samples presented a protein con 1395 µ g/mL, and cancer samples presented a protein concentration o The results show significant differences between groups, with a high centration in cancer samples than the other samples. Finally, these pro by Western blot ( Figure 5).

Discussion
BLF has been demonstrated to have great potential as a research t been used in various diseases, such as for enzyme research on colore endoscopic screening for food allergies [22], CRC diagnosis [19], and m of inflammatory bowel diseases [23]. However, BLF is a rarely used sam considerable potential for research [24] due to its proximity to the in cause this sample type does not cause extra discomfort to patients u copy, since bowel preparation is the same as that required for a colon

Discussion
BLF has been demonstrated to have great potential as a research tool [16], since it has been used in various diseases, such as for enzyme research on colorectal polyps [21], for endoscopic screening for food allergies [22], CRC diagnosis [19], and molecular screening of inflammatory bowel diseases [23]. However, BLF is a rarely used sample with type with considerable potential for research [24] due to its proximity to the injured area and because this sample type does not cause extra discomfort to patients undergoing colonoscopy, since bowel preparation is the same as that required for a colonoscopy [16]. Nowadays, the use of EVs in liquid samples, such as saliva, amniotic fluid, breast milk, seminal liquid, nasal secretion, cerebrospinal fluid, lymph node secretion, urine, plasma, serum, placenta, bronchoalveolar fluid, synovial liquid, bile fluid, and ascites [9,25,26], is widespread, but there are no studies on EVs from BLF.
Despite this lack of studies, our results show that EVs can be isolated from BLF, as demonstrated by AFM and TEM results. Moreover, the NTA results demonstrate that these EVs presented a regular size, considering that exosomes have a diameter of 30-100 nm and microvesicles have a diameter of 100 nm −1 µm [27]. The size difference between TEM and NTA results can be attributed to the dehydrating conditions to which EVs are subjected during fixation for TEM analysis [8], in addition to the presence of a sufficient concentration of EVs in BLF, taking into account that in urine, the EV concentration determined by NTA is 1.00 × 10 10 [28], whereas in CRC patient blood, the EV concentration is 1.29 × 10 9 ± 9.92 × 10 8 [29]. Finally, CD9 and CD63 expression, which are two tetraspanins considered EV biomarkers [8], has been found in EVs isolated from BLF. Altogether, these results suggest that EVs can be isolated from BLF.
EV cargo can be determined and studied, since RNA can be used to study gene expression levels [30], DNA mutations can be determined [31], and protein expression levels can be analyzed by Western blot and liquid chromatography-tandem mass spectrometry [29,[32][33][34]. However, given the nature of our sample, it was necessary to confirm the content of these EVs from BLF as a feasible tool for CRC study. The RNA concentration results presented no differences between groups and demonstrated that the RNA concentration and quality are sufficient to determine gene expression levels, as can be seen in the B2M expression levels, which is an often used housekeeping gene due to its high and constant expression [35]. In addition, the protein concentration results in EVs did not present differences between groups, indicating an adequate concentration and quality for use for determination of protein expression levels, as can be seen in the CD9 and CD63 expression levels. However, the DNA concentration was lower in cancer samples. In contrast to our results, Bryzgunova and collaborators demonstrated that EVs of urine samples from patients with prostate cancer presented higher DNA concentrations than healthy samples [36], although these variances could be attributed to the different type of sample and cancer from which EVs were isolated. Despite this, DNA from EVs can be studied [31], and our results indicate a sufficient concentration of DNA for use, as shown in studies by Thakur and collaborators, who found that DNA concentration in serum EVs was 10.59 ± 13.19 ng/mL [37], or in studies by Choi et al., who determined that the amount of DNA in plasma EVs ranged from 0.1 to 2.48 ng [38]. Finally, protein expression levels can be directly determined in BLF samples, as shown in studies by Kayazawa and collaborators, in which lactoferrin, polymorphonuclear neutrophil elastase, myeloperoxidase, and lysozyme concentrations were determined in BLF samples by ELISA [39]. BLF protein concentrations in cancer samples; however Al-Muhtaseb and Bel'skaya demonstrated that in saliva from breast cancer patients, the total protein concentration was lower than that in control patients [40,41]; however these differences can could be attributed to the distinct sample and cancer types. Nevertheless, BLF protein presented an optimal concentration for use to determine protein expression levels, as shown by the α-Tubulin and GAPDH expression levels results, which are often-used proteins for Western blot analysis due to their high expression [42][43][44], demonstrating that this protein from BLF is suitable for Western blot determinations.

Patients and Sample Acquisition
A total of 170 patients (83 females with an average age 63.6 ± 11.4 and 87 males with an average age 63.8 ± 11.1) were included to carry out the study. Patients of both sexes received the a patient's information sheet about the research project, and they signed an informed consent according to the "World Medical Association Declaration of Helsinki" for medical research involving humans. The study was approved by the Comité d'Ètica de la Investigació de les Illes Balears (IB 3833/19 PI). Days prior to colonoscopy, patients received the information about colonoscopy and an information sheet about the research project. Then, 48 h before colonoscopy, patients were required to reduce fiber, fat, and gas intake, and 8 h before colonoscopy, patients drank 3 L of a polyethylene glycol solution to clean the bowel lumen. Colonoscopies were performed in the endoscopy unit at the Hospital Comarcal d'Inca with anesthetic sedation. BLF samples were recollected with saline solution (0.9% NaCl) applied directly to the injury area in the mucosa through an endoscopic flushing channel, then aspirated and retained in the polyp-trapping basket of the endoscope in order to avoids mixture with fluids from non-affected zones. The samples were stored at −80 • C and divided according to pathology: without pathology (healthy samples; n = 43), with polyps (divided into low-and high-risk of suffering from neoproliferative processes; n = 58 and n = 29, respectively), and cancer samples (n = 40). Low-risk samples correspond to 1 or 2 tubular adenomatous lesions with low-grade dysplasia and serrated lesions without dysplasia-all less than 10 mm; high-risk samples correspond to 3 or more tubular adenomatous lesions with low-grade dysplasia less than 10 mm, at least one adenomatous lesion with a villous component, high-grade dysplasia of more than 10 mm, and at least one serrated lesion with dysplasia or more than 10 mm [45].

Extracellular Vesicles Isolation by Ultracentrifugation
BLF samples were centrifuged for 15 min at 2000× g and 4 • C to eliminate debris and cellular components. Then, 4 mL of supernatant was separated, and 8 mL of sterile PBS 1× (500 mM NaCl, 167 mM NaH 2 PO 4 ·2H 2 O, 333 mM Na 2 HPO 4 ) pH 7.5 was added. Next, this mix was centrifuged at 2000× g for 10 min and 4 • C in order to remove the remaining debris. Finally, supernatants were centrifuged at 2000× g for 10 min at 4 • C to eliminate large particles. Subsequently, supernatants were centrifugated for 1 h at 4 • C and 100,000× g in order to precipitate EVs. The resulting pellets containing EVs were resuspended in 100 µL of sterile PBS 1× pH 7.5.

Atomic Force Microscopy (AFM)
First 50 µL of EV resuspension from healthy and cancer samples was added to a freshly cleaved muscovite mica surface (NanoAndMore GmbH, Wetzlar, Germany) for 10 min, washed with 2.5 mL of deionized water, and dried with nitrogen. Then, EVs were observed under an atomic force microscope (Veeco, Oyster Bay, NY, USA) in tapping mode with aluminum-coated silicon probe tips (HQ:NSC35/Al BS, Mikromasch, Lady's Island, SC, USA). The height and amplitude of the samples were recorded at 512 pixels × 512 pixels at a scanning rate of 1 Hz and processed with NanoScope Image Software (v5.10, Veeco, Metrology, NY, USA).

Nanoparticle Tracking Analysis (NTA)
EV size distribution and particle concentration were analyzed using a Nanosight NS3000 (Malvern Instruments, Malvern, PA, USA). Samples were diluted 1:1000 in a final volume of 1 mL before analysis. Then, samples were passed through the chamber in vivo and recorded three times for 1 min each with a laser at a wavelength of 532 nm and an sCMOS camera. Finally, data were analyzed with NTA 3.2 Dev Build 3.2.16 software.

RNA Isolation, Reverse Transcription, and Real-Time PCR
Total RNA from EVs was isolated using TRI reagent LS ® (T3934, Sigma-Aldrich) following the manufacturer's protocol. Briefly, TRI reagent was added to EV samples and left for 5 min at room temperature. Then, chloroform was added and incubated for 15 min at room temperature, followed by centrifugation at 12,000× g for 15 min at 4 • C. After centrifugation, three phases were differentiated. Isopropanol was added to the aqueous phase and incubated overnight at −20 • C for better RNA precipitation. After the incubation, samples were centrifuged at 12,000× g for 8 min at 4 • C, and the resulting pellets were washed with frozen 75% ethanol and centrifuged at 7500× g for 5 min at 4 • C. The supernatants were discarded, and the pellets were dried under vacuum. Finally, RNA was resuspended in 20 µL of RNase-free water and quantified using a BIO-TEK PowerWave XS spectrophotometer at wavelength of 260 nm. The RNA quality was checked by a 260/280 ratio.
The obtained RNA was mixed to create two or three RNA pools for each sample type to better represent each sample type, avoiding the particular patient characteristics. Then, 300 ng of the total RNA was reverse-transcribed to cDNA. First, a denaturalization at 90 • C for 1 min was applied to RNA samples. Next, the reverse transcription reagents were A LightCycler 480 System II rapid thermal cycler (Roche Diagnostics, Basel, Switzerland) with SYBR Green technology was used to carry out the real-time PCR, following the manufacturer's protocol. The expression of beta-2-microglobulin (B2M) was analyzed (forward primer: 5 -TTT CAT CCA TCC gAC ATT GA-3 ; reverse primer: 5 -Cgg CAg gCA TAC TCA TCT TT-3 ; accession number: NM_004048). The first step in the amplification program was preincubation for cDNA denaturation at 95 • C for 5 min, followed by 50 cycles of denaturation at 95 • C for 10 s, annealing at 54 • C for 10 s, and elongation at 72 • C for 12 s. Finally, melting was applied at 95 • C for 5 s, followed by 65 • C for 1 min, and 97 • C continuously until cooling at 40 • C.

DNA Isolation and Quantification
DNA from EVs was isolated using TRI reagent LS ® (T3934, Sigma-Aldrich) following the manufacturer's protocol. Briefly, after the formation of three phases, 100% ethanol was added to the interphase and organic phase, mixed, and incubated for 3 min at room temperature. Next, samples were centrifuged at 2000× g for 5 min at 4 • C, and the supernatants were saved in a new tube for protein isolation. The pellets were mixed with 0.1 M trisodium citrate in 10% ethanol and incubated for 30 min at room temperature, followed by centrifugation at 2000× g for 5 min at 4 • C; this step was repeated twice. Then, pellets were washed with 75% ethanol and incubated for 20 min at room temperature, followed by centrifugation at 2000× g for 5 min at 4 • C. The resulting pellets were dried at room temperature for 15 min and resuspended in 100 µL of 8 mM NaOH. Finally, centrifugation was performed at 12,000× g for 10 at 4 • C, and supernatants were saved. Finally, DNA was quantified using a BIO-TEK PowerWave XS spectrophotometer at a wavelength of 260 nm. The DNA quality was checked by a 260/280 ratio.

Protein Isolation and Quantification
Protein from EVs was isolated using TRI reagent LS ® (T3934, Sigma-Aldrich) following the manufacturer's protocol. Briefly, the supernatants saved during DNA isolation were incubated with isopropanol at room temperature for 10 min and centrifuged at 12,000× g for 10 min at 4 • C. The supernatants were discarded, and precipitates were washed three times with 0.3 M guanidine hydrochloride in 95% ethanol, incubated at room temperature for 20 min, and centrifuged at 7500× g for 5 min at 4 • C. After three washes, 100% ethanol was added to the precipitates and incubated at room temperature for 20 min, followed by centrifugation at 7500× g for 5 min at 4 • C. Next, protein pellets were air-dried for 15 min, dissolved in 100 µL of 1% SDS, and incubated overnight at −20 • C. Then, samples were centrifuged at 10,000× g for 10 min at 4 • C, and the supernatant was transferred to a new tube. Finally, protein was quantified by a Pierce ® BCA protein assay kit (10741395, Fisher Scientific) following the manufacturer's protocol.
Protein from BLF samples was directly quantified by the Bradford method [47].

Statistical Analysis
Statistical Program for the Social Sciences software for Windows (SPSS, version 25.0; SPSS Inc., Chicago, IL, USA) was used to perform all statistical analyses. First, a boxplot was used to discard the outliers. Then, a normality study was performed using the Shapiro-Wilk test; for parametric results (EV size, EV concentration, and B2M Ct values) a one-way ANOVA was used to analyze differences between groups; for non-parametric results (RNA, DNA, EV and BLF protein concentrations, and B2M Tm values), the Kruskal-Wallis test was used to analyze differences between groups. All determinations were made with minimal statistical significance at p < 0.05, and all results are presented as mean ± SD.

Conclusions
In conclusion, bowel lavage fluid is a sample that must be taken into account in the study of colorectal cancer due to its proximity to the tumor. In addition, the extracellular vesicles isolated from this sample type can be useful as a source of colorectal cancer biomarkers, considering that EV content can be determined and studied by different molecular biology techniques. The possibility of studying the content of extracellular vesicles isolated from bowel lavage fluid could improve knowledge of colorectal cancer, in addition to identifying new biomarkers (for diagnosis, prognosis, and monitoring of the disease), which could be extrapolated to non-invasive samples, such as stool samples. Nevertheless, further investigation of bowel lavage fluid and, specifically the extracellular vesicles from such samples, is necessary to better understand the mechanism whereby EVs are released from cancer cells and the role that their content plays in cancer initiation, progression, and metastasis.