Overexpression of Argonaute HIWI Gene in Colorectal Cancer Stem Cells and Colorectal Cancerous Tissue

Background: Recent studies have shown that Hiwi has a crucial role in stem cell self-renewal in various organisms and it is also associated with some cancers. In the present study, Hiwi expression was examined in different grades of primary human colorectal cancer (CRC), colorectal cancer stem cells (CRCSCs) and HT-29 CRC cell line. Methods and Results: CRC tissue samples were collected from 20 patients with CRC. Furthermore, HT-29 cell line, CRCSC and 13 normal colorectal tissue samples were prepared. Expression of Hiwi at mRNA and protein levels was determined by Real-time PCR, ow cytometry, Western blot and immunohistochemistry. Overexpression of Hiwi was detected in 40% (3.351 ± 2.94, P<0.05) of clinical CRC tissue specimens. We also observed a signicant increase in the expression of Hiwi in CRCSCs (5.94 ± 0.05, P < 0.005). Hiwi expression in CRCSC was signicantly higher compared to the colorectal cancer tissue and HT-29 cells (p<0.01). Hiwi mRNA level was signicantly correlated to tumor grade (p < 0.01) and stage (p < 0.01). Conclusions: HiWi can be considered as an oncogene in progressive cancer and, therefore, it can be a valuable biomarker and target in therapeutic procedures.


Introduction
Colorectal cancer (CRC) is the third most common cancer and the second most common cause of cancerrelated death in both men and women [1][2][3]. Although multiple factors are responsible for colorectal cancer development, it is mainly resulted from genetic and epigenetic variations which convert normal glandular epithelium into invasive adenocarcinoma. At present, multiple cellular and molecular pathways have been proposed for the development of colorectal cancer [1,4]. Identi cation of molecules or associated risk factors involved in colorectal cancer development is valuable for early diagnosis and also nding a better therapeutic strategy for cancer at early stages [4,5].
Cancer stem cells (CSCs) are a rare population of cells within a tumor which serve as tumor-initiating cells or tumorigenic cells [6,7]. CSCs have a similar expression of cell surface markers pattern like normal stem cells. Furthermore, They have self-renewing capacity and can be differentiated to various cancer cells type with common signaling pathways [6]. Unlike normal stem cells, CSCs have a tumorigenic activity that enables them to form tumors when they are transplanted into animal models [7].
Piwi like RNA-mediated gene silencing 1 (HIWI), also known as Piwi1, is a member of the highly conserved P-element induced wimpy testis (PIWI) family. are The Piwi family are important regulators of different physiological processes such as germ cell differentiation, stem cell self-renewal and differentiation, spermatogenesis and male infertility [8,9]. Expression of Hiwi protein in spermatocytes and round spermatids during spermatogenesis indicates that this gene is critical for normal spermatogenesis.
Most of Hiwi mRNAs are not complementary to their target genes, which suggest that Hiwi family are more likely under the effect of post-transcriptional regulation rather than chromatin-associated regulation. It has been determined that Hiwi are exclusively expressed in stem cells. For example, Sharma et al., [10] found HIWI expression in CD34 + hematopoietic stem cells (HSCs), but not in differentiated HSCs. This data indicates that HIWI may be responsible for determination or regulation of HSCs development.
Recent studies have shown that HIWI is ectopically expressed in all cancers [11][12][13]. It has been shown to be up-regulated in different cancers such as seminomas [11], gliomas [14], squamous-cell carcinomas [15], pancreatic [16], liver [17], and gastric cancers [18]. Taubert et al., have reported that Hiwi is highly expressed in primary soft-tissue sarcomas, where higher Hiwi mRNA levels are predictable based on fault clinical outcomes [12,13]. However, the exact function of Hiwi in tumorigenesis is unclear. Araújo et al., [19] demonstrated that Hiwi up-regulation plays a crucial role in the signaling pathway of gastric cancer and promotes CSCs maintenance, tumor cell viability, migration and invasion. A previous study found overexpression of Hiwi in testicular seminoma, indicating Hiwi deregulation may contribute to the occurrence and development of testicular tumors [20]. These data mean that abnormal expression of Hiwi protein may be involved in development, progression and poorer diagnosis of various cancer forms [12,13]. On the other hand, some studies found reverse association between Hiwi expression and cancer progression. For example, Wang et al., [21] revealed that overexpression of Hiwi inhibits the growth of chronic myeloid leukemia K562 cells and consequently enhance their chemosensitivity to daunomycin.
These data implicate that the biological functions of Hiwi may vary between different types of tumor, necessitating its role in each cancer to be studied individually. Furthermore, Hiwi may be considered as a potential target for cancer diagnosis and therapy because most non-cancer cells would not be affected by cytotoxic effects [19]. Given the critical role of Hiwi in cancer proliferation and metastasis, we assume that altered expression of Hiwi may be involved in development of colorectal cancer. Therefore, this study aims to evaluate the Hiwi expression and its prognostic values in 20 clinical samples with non-necrotic colorectal cancer tissue and normal respective colorectal tissue. Furthermore, Hiwi expression levels are compared between different grades of primary human colorectal carcinoma, colorectal cancer stem cells (CRCSCs), as well as HT-29 CRC cell line.

Methods And Materials
Study population and sample collection A total number of 20 patients (aged 21-46 years old) with colorectal cancer and 13 healthy individuals who referred to Imam Khomeini Hospital (Tehran-Iran), were entered into the study between 2016 and 2018. Healthy controls were those who came to our hospital for checkup and no abnormalities were found in physical examination or laboratory results. They also did not have a history of cancer and previous medical diseases. The control subjects were matched to the patients based on age and sex. This cross-sectional study was approved by the institutional review board and ethics committee of Tehran University of Medical Sciences (Tehran, Iran). Written informed consents were signed by all subjects at the department of cancer in Imam Khomeini Hospital. Clinical and pathological ndings of the patients, including disease stage, tumor grading, metastasis status, and lymph node involvement. were also provided. Inclusion criteria for patients were: (i) patients with known colorectal cancer; (ii) complete clinical and pathologic information. Patients who met the following criteria were excluded from the study: (i) the presence of tumor in other places; (ii) previous treatments such as chemotherapy or radiotherapy; (iii) history of other chronic diseases like diabetes mellitus, and liver disease.
All colorectal cancer tissue specimens were con rmed by the surgical ndings and the postoperative pathological study. All specimens were collected and transferred into RNAlater Reagent (Qiagen, USA) and kept at 4°C. The tissue samples were washed with phosphate-buffered saline (PBS) and cut into smaller pieces and stored at −80°C until use.

Cell separation by CD44 positive selection
The human colorectal adenocarcinoma cell line HT-29 was passed through a MACS column [Order number: 130-095-194, Miltenyi Biotec, UK], including CD44 monoclonal antibody, thereby CD44 + HT29 cells were separated from other cells, and then they were cultured. The isolation of cells was performed according to the method described by Gao et al. [22]. Quantitative RT-PCR Total RNA was extracted using Qiagen RNase kit. Complimentary DNA (cDNA) was synthesized using 1 µg of RNA, according to instructions in the cDNA synthesis kit (Takara, Japan). Real-time PCR was performed by SYBR green PCR Master Mix (Takara), on a Rotor-Gene Q, real-time thermocycler (Qiagen, USA) and Real-time RT-PCR primers for Hiwi and GAPDH were designed by Allele ID software Version 6.0 (PREMIER Biosoft, CA, USA) ( Table 1). The thermal pro le included 10 min at 95°C followed by 40 cycles of 15 sec at 95°C, 30 sec at 57°C, and 45 sec at 72°C. Data were normalized to GAPDH expression applying the comparative threshold cycle method. The PCR e ciencies for Hiwi and GAPDH were veri ed by generating related standard curves. The relative expression of Hiwi was compared based on uorescence intensity changes of samples from endometriosis vs. compliant normal tissues. More than two-fold increase in expression was considered as over-expression, while more than two-fold decrease was considered as under-expression. The range between those two values was interpreted as no change or regular expression. All experiments were performed in triplicate. Immunohistochemistry Immunohistochemistry (IHC) was performed as previously described by Taubert et al. [11,12]. Brie y, para n-embedded tissue sections were depara nized and rehydrated. Slides were treated with 3% H 2 O 2 for 10 min and 2% BSA for 30 min at room temperature before incubation overnight with 2 µg/ml polyclonal anti-PIWI (HIWI) (abcam, UK) antibody at 4 C. After 3 washes in PBS, slides were incubated with an HRP-linked anti-mouse secondary antibody for 30 min, and then washed 3 times with PBS, followed by chromogen detection with DAB for 10 min, and hematoxylin counterstaining.

Western blot analysis
Samples were collected and lysed with standard RIPA buffer. After centrifugation at 40,000 g at 4 o C for 45 min, the supernatant was subjected to SDS-PAGE analysis. 100 µg of protein from each clinical sample, CD44+ HT29 and HT-29 colon cancer cell line was subjected to 12% SDS-PAGE. Proteins were transferred to the nitrocellulose membrane (Millipore, USA) using Western blot technique followed by blocking using 5% bovine serum albumin (BSA, Sigma, Germany) for 60 min at room temperature along with shaking. Nitrocellulose membranes were then incubated with rabbit anti-PIWI (abcam, UK) antibody (1:1000) for 1 h followed by the incubation of membranes with HRP conjugated anti-mouse antibody at room temperature with mild shaking. Enhanced chemiluminescence (ECL) Western blotting system (GE Healthcare, USA) was used to develop the membrane on high performance chemiluminescence lm (GE Healthcare) according to the company guidelines. After each step, the membrane was washed with PBS [11,13].

Statistical analysis
Data were analyzed using the SPSS software. Pearson's correlation assessed correlation between HIWI expression levels in various cancer stages and grades. Comparison of the gene expression between all groups was evaluated by ANOVA, Tukey test analysis. P value <0.05 was considered to be statistically signi cant.

Cancer stem cell isolation and identi cation
Colorectal cancer stem cells (CD44 + ) were successfully isolated from HT-29 colorectal adenocarcinoma cells line after positive immune-selection to deplete CD44 cells ( Figure 1). As shown in gure 1, the morphological appearance of colorectal cancer stem cells (CD44 + ) is different from HT-29 cells line.
Colorectal cancer stem cells (CD44 + ) appeared as cuboidal-shaped cells with scant cytoplasm and granules around the nuclei (Figure. 1B). In compared to the HT-29 cell line ( Figure. 1D) derived from the adenocarcinoma patients, they have a same proliferation capacity. Flow cytometry analysis revealed that colorectal cancer stem cells were strongly positive for surface markers CD44, CD133, CD90, but they were negative for CD45 and CD34.

Real time PCR analysis of Hiwi Gene
The relative expression of Hiwi in CD44 positive colorectal cancer cells, HT-29 cell line, and normal colorectal cells was analyzed by Real time PCR. As shown in Figure 2, the relative expression of Hiwi gene was higher in colorectal adenocarcinoma samples and CD44 positive colorectal cancer stem cells compared to normal cells. This result indicates that Hiwi gene is down-regulated in normal colon samples.

Immunohistochemistry (IHC) staining
Hematoxylin & Eosin staining of normal tissues ( Figure 3A) and colorectal cancer ( Figure 3B) immunohistochemically for HIWI was localized in the colon of cancerous patients ( Figure 3C). IHC analysis showed increased expression of HiWi protein in colorectal tissues compared to normal colon tissues.

Western blot analysis
Western blot analysis was performed to con rm the expression of HIWI protein in colorectal tissue extract. Expression of HIWIprotein was observed under western blot analysis in tissue extract ( Figure 4).

Hiwi mRNA expression in different disease stages and grades
The expression levels of Hiwi mRNA were normalized with respective mRNA levels of GAPDH as a housekeeping gene. Mean expression of Hiwi mRNA in colorectal tumor tissue was signi cantly higher than healthy colon sample (p<0.001). There were statistically signi cant differences in Hiwi expression in patient's group according to the disease stage and grade ( Figure 5). Ten patients (50%) were in stage I/II, and the others (50%) were in stage III and IV ( Table 2). The mean expression of Hiwi mRNA levels in patients with disease stage I/II was signi cantly higher than that in patients with disease stage III/IV (p<0.001; Fig. 5). The relative expressions of Hiwi gene in patients with early and advanced stages of colorectal cancer were 3.92 and 1.58, respectively ( Figure 5).
There was a signi cant difference in the mean expression of Hiwi mRNA levels between patients with different disease grades (p<0.01). Ten patients (50%) were in grade I, 5 patients (25%) in grade II and 5 patients (25%) in grade III ( Table 2). The mean expression of Hiwi mRNA levels in grade I was signi cantly higher than that in patients with grade II and III (p<0.01; Fig. 5).

Discussion
Due to the high incidence of colorectal cancer and its adverse effects on social life, it is essential to identify and target molecules or factors that are involved in cancer development and progression. Cancer cells have similar proliferation and differentiation patterns to stem cells. Therefore, it is important to study the genes/proteins that are responsible cancer stem cells differentiation to various cancer types. Recent studies have proposed that HIWI plays a crucial role in cancer development and progression. Therefore, it can be considered as a valuable target for cancer treatment. In the current study, we considered the expression pattern of Hiwi at both protein and mRNA levels in CD44 positive colorectal cancer stem cells, HT-29 cell line, healthy and cancerous colorectal tissues via real-time quantitative PCR, western blot assay, and immunohistochemical staining methods. We found overexpression of Hiwi in 40% of clinical colorectal cancer specimens. Similarly, Raeisossadati et al., [23] reported overexpression of Hiwi gene in 34.8% of clinical colorectal cancer specimens. In the study by Zeng et al., [24], they found overexpression of HIWI protein in 25.6% of patients with colorectal cancer. These ndings are comparable with results of our study. We also found that relative expression of Hiwi at mRNA and protein levels was signi cantly higher in patients with colorectal cancer than healthy individuals. Interestingly, this expression pattern in CCSC was even higher than in cancerous colorectal tissues and colorectal cancer cells line. More importantly, the Hiwi expression pro le was signi cantly correlated to the disease stages and grades. Patients with diseases stage I/II and grade I showed signi cantly higher Hiwi expression level compared to those with disease stage III/IV and grades II and III. These data indicate that Hiwi may be involved in colorectal cancer development and progression and, therefore, it can be considered as a potential prognostic biomarker for patients with colorectal cancer, especially at early stages.
To support our ndings, a growing number of studies reported overexpression of Hiwi in different types of cancers [12,13]. For example, Zeng et al., [24] found that patients with colorectal cancer and positive Hiwi expression had signi cantly lower survival rate compared to patients with negative Hiwi expression.
Litwin et al., [25] found increased expression of Hiwi mRNA in colorectal cancer tissues compared to noncancerous samples, which is in accordance with the ndings of our study. Some studies indicated that up-regulation of Hiwi is associated with clinical features of cancerous patients [12,13,26]. Similarly, we found a signi cant relationship between colorectal cancer disease stages and degrees with Hiwi expression pattern. Our data also revealed that Hiwi is up-regulated in colorectal cancer cells and its function is altered in aberrant cancer state. Therefore, Hiwi can be considered as a direct tumorigenic or oncogenic factor. To support this hypothesis, Raeisossadati et al., [23] considered Hiwi as a tumorigenic factor in colorectal cancer. They showed that increased expression of Hiwi in colorectal cancer cells was signi cantly associated with the depth of tumor invasion, the stage of tumorigenesis progression and lymph node metastasis of tumor cells [23]. Wang et al., [27] reported overexpression of Hiwi in both clinical breast cancer specimens and breast cancer cell's lines. Additionally, they found that increased expression of Hiwi was signi cantly associated with the growth of human breast cancer cells, tumor size, lymph node metastasis and histological grade [27]. A previous study reported increased expression of Hiwi in the cytoplasm of esophageal cancer cells which was associated with histological grade, disease stage and clinical outcome [28]. We also found increased expression of HIWI protein in undifferentiated human colorectal sarcomas; however, the exact molecular mechanism for this up-regulation in sarcomas is unclear. In a previous study, Liu et al.,[29] found overexpression of Hiwi in human gastric cancer which was associated with the proliferation of cancer cells. A recent study has revealed that Hiwi promotes proliferation of colorectal cancer cells through upregulation of global DNA methylation [30].
According to the previous accomplished data and ndings of our study, Hiwi can be considered as an oncogene for colorectal cancer development and progression. Therefore, it can be targeted for colorectal cancer therapy, especially at early stages.

Conclusion
In summary, our results showed that Hiwi expression at both mRNA and protein levels is signi cantly increased in clinical colorectal cancer tissues compared to normal specimens. Furthermore, the expression of Hiwi in CCSC was signi cantly higher than in colorectal cancer tissues than normal specimens. We also found a close relationship between Hiwi expression and colorectal cancer stage and degree. These results suggest that Hiwi may be involved in the development and progression of colorectal cancer and can be considered as a prognosis factor for the early diagnosis of this cancer. Besides, the oncogenic role of Hiwi indicates that it may be a potential target for cancer therapy. Further studies are merit to discover the underlying mechanisms by which Hiwi promotes colorectal cancer development.     Overexpression of HIWI protein in colorectal cancer was detected by Western blotting. GAPDH was used as a loading control. S1 and S2, cancer samples (n = 2); S3, CD44 positive CSCs; S4, HT-29 colorectal cancer cell line.

Figure 5
Ratios of Hiwi mRNA expression in colorectal cancer, correlated best with the corresponding ratios normalized for GAPDH (R = 0.674; P < 0.001). Ratios of Hiwi mRNA expression were not associated with gender (men vs. women, P = 1.0), tumor grade (P = 0.0667) or stage (P = 0.0514).