Tumor-exosomal miR-205-5p as a diagnostic biomarker for colorectal cancer

Background: Tumor-exosomal miRNA play crucial roles in cancer diagnosis. The current reports aimed to found an exosomal miRNA for colorectal cancer(CRC) with non-invasiveness, sensitivity and speci�city. Patients and methods: The exosome was extracted from CRC patients and healthy donors using ultracentrifugation to verify by TEM, qNano and western blot. Differential expression level and clinical characterization of miR-205-5p were analyzed in colorectal cancer via TCGA. Real Time Quantitative PCR was employed to determine the different expression levels of exosomal miRNAs in 157 primary CRC patients and 135 healthy donors. Predictions were made concerning target genes to determine the direction for further exploring the etiopathogenesis of the disease by means of bioinformatics. Results: The expression of miR-205-5p demonstrated a substantial downregulation in colorectal cancer compared to healthy donors, as determined through analysis of the TCGA database. We conducted a prediction and analysis of the functional enrichment of downstream target genes regulated by miR-205-5p. Reduced level of exosomal miR-205-5p in serum from CRC patients was found compared with healthy controls (p<0.0001, respectively) and benign disease (p<0.0001, respectively). The levels of expression of exosomal miR-205-5p were substantially lower in early stage CRC patients than in the contrast groups (p<0.0001, respectively). The expression levels of exosomal miR-205-5p saw signi�cant increase postoperatively (p=0.0053, respectively). Conclusions: The present study demonstrated that serum exosomal miR-205-5p may be a diagnostic biomarker for colorectal cancer.


Introduction
Colorectal cancer (CRC) ranks third among the commonest cancers and poses a big threat to people's health and life.Nevertheless, approximately four out of ten CRC patients end up with relapse and recurrent metastasis, and merely a small portion of them have a survival time of ve years (1,2).Hence, the development of effective and easy-to-use screening tools is of vital importance for early diagnosis and precise stage classi cation in CRC patients.These may include uid biopsy such as proteins, tumor markers, circulating tumor cells (3), cell-free DNA(4), and exosomes (5), to facilitate the early detection and precise stage classi cation of CRC(6).
Exosomes, also referred to as extracellular vesicles, are secreted organelles with a single membrane and measure 30 to 200 nm in diameter (7).Cells communicate via exosomes (8).Therefore, as an essential messenger between cells, exosomes transmit signals to neighbouring cells or to distant anatomical sites by carrying cytokines (9), modify proteic and genetic expression in receiving cells, and thus in uence the functions of receiving cells.Exosomes play multi-roles as well in the pathogenesis (10) metastasis (11) and immunity (12).Exosomes encompass a variety of proteins (13), miRNAs (14) and DNA (15), which are apt to be transferred to target cells.For instance, extensive deregulation of miRNAs have shown in human cancers, which highlights their key role in the onset, development and metastasis of the tumor, and hence miRNAs become potential biomarkers for cancers.
MicroRNA (miRNA) are tiny, non-coding RNA which are generally 20-25 nucleotide long sequences(16), and which will unite with the target mRNA sequence after they enter the receptor cell (17).A host of studies have demonstrated that exosomal miRNA is capable of mediating communication between cells, thus involved in tumor carcinogenesis, metastasis, immunosuppression, angiogenesis, and other aspects (18,19).Emerging evidence suggested that ve types of exosomal miRNAs, namely miR-205, miR-19a, miR-19b, miR-30b and miR-20a, are promising diagnostic markers of squamous cell lung cancer due to their decreased levels of circulation after resection operation of the lesion (20).The level of serum exosomal miR-205 was up-regulated in ovarian cancer tissues, and its over-expression was linked to cancerous metastasis in ovarian carcinoma patients (11).The relative expressions of exosomal miR-1246 in serum were signifcantly overexpressed in gastric cancer patients compared to that in healthy controls (21).Nasopharyngeal carcinoma cells discharge exosomes that contain certain types of miRNAs that play an inhibitory role in T-cell proliferation,targeting MAPK-1 and STAT pathways (22).
The increase of neutrophil count in peripheral blood has been shown to be related to poor clinical outcomes in pancreatic cancer, gastric cancer and breast cancer.Rao et al demonstrated that the increase in intertumoral neutrophils was associated with malignant phenotypes and could predict adverse prognosis in CRC.
This study aimed to verify the different expression levels of serum exosomal miRNA in CRC patients and in contrast groups by means of small RNA sequencing and uorescence quantitative PCR.Consequently, exosomal miR-205-5p were chosen and their related with diagnostic e ciency and clinical features was analyzed.Therefore, exosomal miR-205-5p as a novel diagnostic biomarker.

Patients and clinical samples
In the study, we selected altogether 157 CRC patients, 135 healthy donors and 20 benign disease patients hospitalized at the Shangdong Caner Hospital in September 2017 to July 2018.We obtained their written consent.The estimation of tumor staging was performed in accordance with the AJCC Cancer Staging Handbook, 2017.The protocol was approved by the Shandong Cancer Hospital A liated to Shandong First Medical University and Shandong Academy of Medical Sciences of committee.All subjects gave written informed consent in accordance with the Declaration of Helsinki.No patient was given anti-tumor treatment prior to serum collection and no patient was diagnosed with diseases of the endocrine, immune or metabolic system.Sera collection was performed on 17/157 patients with CRC, who received surgical treatment after two months.Patients' clinical features and diabetic history were investigated (Table 1).(23) under the condition of 10000 ×g for 30 min at 4°C for the purpose of removing the cellular debris, and then under the condition of 100,000 ×g for 2 h at 4°C for exosome precipitation.Then, the samples were analyzed using the following techniques.

TEM assay
Transmission electron microscopy (TEM) was conducted for the purpose of identifying the puri ed exosomes.The exosome pellets were moved to the grids of solution (a 50 µL drop of 1% glutaraldehyde), and after a standing time of 5 minutes removed into distilled water (100-µL drop), and after another standing time of 2 minutes, a 50-µL drop of uranyl-oxalate solution (Ph = 7) was applied to the grids for 5 minutes and covered according to instructions.Afterwards, the grids were washed with distilled water, 2 minutes for each, and examined under a TEM machine.

Tunable resistive pulse sensing (TRPS)
The measurement of nanoparticle size was conducted with the TRPS technique and on the qNano ( Izon Science Ltd).Data analysis was conducted through Izon Control Suite v.3.3.2.2000 (ibid).

Immunoblotting
After being resolved by SDS-PAGE, exosomal proteins of equal amounts were removed to PVDF membranes, which were treated with a solution (5% milk in Tris-buffered saline containing 0.1% Tween 20) for 1 hour at 4°C in dark environments with rabbit primary antibodies against CD81, TSG101, and GM130.Afterwards, they were incubated with HRP-conjugated secondary antibodies for 1 hour at normal room temperature.ECL blotting detection reagents were eventually applied.

Target genes and pathway
TCGA (Https://portal.gdc.cancer.gov)2.7 RNA isolation and real-time PCR TRIzol reagent was selected to extract aggregate RNAs, which were then reversely-transcribed into complementary DNA (cDNA) with the Mix-X miRNA First-Strand Synthesis Kit.Real-time PCR was conducted using TB-Green Premix Ex Taq II Reagent.U6 was viewed as an internal control (24).All these procedures were performed in accordance with the speci c manufacturer's instructions.For each sample, analysis was made in duplicate.In addition, the evaluation of PCR reaction was by means of melting curves and that of relative quanti cation of miRNA expression was through the ΔCT method (CtmiRNA-CtU6) as delineated in the previous section (25).

Statistical analysis
For statistical analysis, SPSS 22.0 and GraphPad Prism 6.0 were selected.Mann-Whitney U or t-test were used for comparison and a paired t-test was performed to compare paired values.As for multiple comparisons, one-way ANOVA was performed.The corresponding cutoff points were determined through ROC curves, where p-value < 0.05 was considered to be of statistical signi cance.Plus, miRNA or miRNAs in combination were determined after analysis.

Characterization of isolated exosomes
Serum exosomes from healthy donors and CRC patients were identi ed using TEM, qNano and immunoblotting.The microvesicles obtained were cup-shaped, under 150 nm in diameter, consistent with exosomal morphology by TEM (Fig. 1A).The qNano analysis was employed to quantify the diameter of a single particle, which was detected to be rounded particle with a diameter of 30 to 150 nm (Fig. 1B).As shown in Fig. 1C, TSG101 and CD81 were positive under Western blotting, mainly were enriched in exosomes.GM130 is the matrix protein regulating the structure of GA, hence generally employed as a negative control for exosome(26), which found close expression in the cell.The above evidence showed that the small vesicles centrifuged from serum were exosomes considering their size and expression level of marker protein.

Identi cation of miR-205-5p in CRC
To verify the differential expression of miR-205-5p in CRC, we conducted an analysis of the TCGA database.The results revealed a signi cant down-regulation of miR-205-5p in CRC compared to healthy donors (Fig. 3A), as illustrated in the Fig. 3B, the AUC of miR-205-5p was 0.896.Adjacent tissues versus cancer tissues from GSE49246, discovered that miR-205-5p was remarkably down-regulated in cancer tissues (Fig. 3C).
To investigate whether miR-205-5p was diagnostic of early CRC via TCGA database, we conducted a further analysis of the differential expression between stage I CRC patients and healthy donors (Fig. 3D), revealing a statistically signi cant difference.The AUC was calculated to be 0.818(Fig.3E).
After conducting an analysis on the relationship between miR-205-5p and clinical characteristics using TCGA data, it was observed that miR-205-5p showed no correlation with age, CEA, or metastasis(Figure 4A,4B and 4F).Nevertheless, it exhibited signi cant variation when considering different BMI index, pathologic T-stage, and pathologic N-stage(Figure 4C-E).Speci cally, miR-205-5p displayed signi cant difference in pathologic staging between + and + (Fig. 4G).Furthermore, it was observed that patients with colon polyps had signi cantly higher expression levels of miR-205-5p compared to those without polyps (Fig. 4H).Additionally, a lower expression of miR-205-5p was found in patients with progressive disease (PD)and stable disease(SD)compared to those with partial response (PR) + complete response (CR) (Fig. 4I).Considering the aforementioned ndings, it can be inferred that miR-205-5p has the potential to serve as a biomarker for the early diagnosis and prognosis of colorectal cancer.

Enrichment analysis of miR-205-5p in CRC
It was predicted that miR-205-5p was involved in the genetic and pathway regulation, providing a direction for exploring the pathogenesis of CRC through bioinformatics.To investigate the target genes of miR-205-5p, as shown in the Fig. 5A, we performed correlation analysis of miR-205-5p with target genes from the TCGA CRC database.Heatmap illustrating the top 50 target genes negatively associated with miR-205-5p (Fig. 5B).The Venn diagram illustrates the overlapping set of predicted target genes of miR-205-5p, target genes that exhibit a negative correlation with miR-205-5p, and differentially expressed genes (DEG) identi ed between healthy donors and patients with CRC retrieved from the TCGA database (Fig. 5C).We performed Gene Ontology (GO) analysis and observed that miR-205-5p exhibited signi cant enrichment in various cellular processes.Speci cally, the analysis revealed enrichment in golgi vesicle transport, endosomal transport, vesicle tethering complex, vesicle-mediated transport to the plasma membrane, and cadherin binding involved in cell-cell adhesion, etc (Fig. 5D).Furthermore, miR-205-5p was found to be predominantly enriched in microRNAs in cancer, colorectal cancer, PI3K-AKT signaling pathway, the thyroid hormone signaling pathway and platinum drug resistance, etc (Fig. 5E).Importantly, we also conducted an analysis of the Hallmark pathway, which exhibited notable enrichment in E2F targets, G2M checkpoint, mitotic spindle and other related pathways (Fig. 5F).The above research results pointed to a fact that miR-205-5p is associated with cancerous onset and progression of cancer.

Characterization of Identi ed Serum Exosomal miR-205-5p.
The expression of miR-205-5p in EDS and exosomes was examined.In effect, miR-205-5p expression in exosomes substantially increased in comparison with EDS (Fig. 6A).What's more, miR-205-5p expression in exosomes underwent no substantial changes after RNase A treatment (Fig. 6B), which proved the stability of exosomal miR-205-5p.Put it simply, the data demonstrated that miR-205-5p chie y existed in the exosomes, serving the role of protecting miRNAs from being degraded by RNases.Notably, exosomes underwent no considerable change in terms of miR-205-5p expression after being exposed to room temperature for a duration of 0, 6, 12, 18, and 24 hours (Fig. 6C).

Exosomal miR-205-5p were associated with CRC
To explore the potential value of exosomal miRNA for cancer diagnosis, exosomal miR-205-5p expression levels were quanti ed in a total 157 CRC patients, 135 healthy donors and 20 benign disease by using RT-qPCR.The results in Fig. 7A suggested that the measured indicator in the serum decreases considerably in the CRC patients(p < 0.0001, respectively)in contrast with healthy donors and benign disease(p < 0.0001, respectively), but not as signi cant as miRNAs were detected among healthy subjects and benign patients (p = 0.184, respectively).
Additionally, we investigated the correlation between miR-205-5p clinicopathological characterist in CRC patients and healthy donors (Table 1).The outcomes suggested that they were not associated with age, gender, histology, progression stage, and among others.
Generally, ideal clinical diagnostic biomarkers should have excellent sensitivity and speci city.In order to test whether serum exosomal miR-205-5p was an ideal diagnostic marker for CRC, ROC curve analysis was performed to determine its expression level.We found the area under the ROC curve (AUC) of 0.639 (95% CI: 0.576-0.702,p < 0.0001, Fig. 7B), and the sensitivity and speci city of detection was 53.5% and 69.2%.
Furthermore, the diagnostic e cacy was evaluated.As is indicated in Fig. 7C, early-stage CRC patients showed a lower level of exosomal miR-205-5p than healthy donors and benign disease (p < 0.001 and p < 0.0001, respectively).Next, ROC curves were devised for the evaluation of exosomal miR-205-5p performance in early diagnosis of CRC.As shown in Fig. 7D, the AUC of miR-205-5p was 0.66 (95% CI: 0.58-0.739)and the sensitivity and speci city were 53.6% and 71.9% respectively.
we continued to observe the expression of the molecules before and after surgical resection in 17 CRC patients.We found that the observed indicator substantially increased after the lesion was removed (p = 0.0053, respectively, Fig. 7E).The above ndings suggested that tumor retention was likely to exert a negative impact on the observed indicator, thereby making it a novel biomarker for surgical e ciency monitoring.

Discussion
Colorectal cancer (CRC) ranks high among the commonest malignant cancers around the globe and poses a big threat to people's health and life (29).Hence, it is of great importance to control the progression of the disease at an early stage.Early diagnosis in CRC patients could realize a ve-year survival for 90% of the cases (30).Yet, delayed diagnosis occurs to nearly 60% of CRC patients, which means a survival rate of merely 8-9%.Hence, a sensitive and speci c biomarker is an urgent demand for distinguishing patients with colorectal cancer.The outcome of this study indicated that exosomal miR-205-5p considerably downregulated in CRC, making them potential biomarkers for CRC diagnosis.
As has been reported, miR-205-5p is involved in the development of various types of cancers (31)(32)(33).For example, overexpression of miR-205-5p was detected in patients with non-small cell lung cancer in comparison with the contrast groups (34), miR-205-5p was signi cantly downregulated in estrogen receptor-positive breast cancer by targeting to NFIB (35).Related studies have shown the regulatory role of miR-205-5p in epithelial-mesenchymal transition by targeting PTEN via PI3K/AKT signaling pathway in cisplatin-resistant nasopharyngeal carcinoma cells (36), miR-205-5p inhibits PTK7, thereby involved in the proliferation, migration and invasion of CRC (37).Similarly, miR-205-5p was modulated by lncRNA NEAT1 to promote CRC cell proliferation and migration through regulating VEGFA signaling pathway (38).
In addition, there is an article report that anti-correlation existed between the expression level of miR-205-5p and BRCA1 and RAD17 targets in HNSCC (39).Therefore, miR-205-59 might be used as putative effective biomarkers for CRC diagnosis.
This study validated that a series of evidence that exosomal miR-205-5p might be employed as diagnostic biomarkers for CRC.miR-205-5p exhibited signi cant downregulation in colorectal cancer compared to healthy donors via TCGA database.exosomal miR-205-5p witnessed considerable downregulation in CRC patients compared to the control subjects.The AUC of exosmal miR-205-5p was 0.639 with the sensitivity of 53.6% and the speci city of 71.3%.The AUC of exosmal miR-205-5p was 0.66 with the sensitivity of 53.6% and the speci city of 71.9% in early stage CRC patients serum exosomal miR-205-5p.It is signi cantly increased after surgery, indicating it is closely related with tumor occupying.Finally, miR-205-5p mainly existed in serum exosome and not affected by RNase A. Apart from that, the target gene of exosomal miR-205-5p and pathway were predicted by bioinformatics, we found exosomal miR-205-5p had close correlation with cancer.
However, the limitations of this study should be pointed out.First, small cohort size did not render su cient statistic support to the conclusion.In future study, a larger sample size and clinical follow-up for a longer term are expected.Second, we did not take into account the diagnostic e cacy of the exosomal miR-205-5p in combination with the routine tumor biomarkers of CRC, like CEA and CA199, since relative data concerning healthy donors were not collected.We intend to make further exploration of serum exosomal miR-205-5p in terms of mechanism and prognosis.
In a nutshell, this study revealed that the expression levels of exosomal

Figures Figure 1
Figures

Figure 4 The
Figure 4

Figure 5 Functional
Figure 5

Table 1
is of practical signi cance for CRC treatment.Therefore, this study provided evidence that exosomal miR-205-5p is a promising diagnostic biomarker for CRC in clinical practice.
miR-205-5p were remarkably lower in CRC patients, including those at the early stage CRC patients.This nding is valuable for CRC diagnosis and