Basically, CIN is characterized by any chromosomal copy number or structure change. CIN is the commonest genomic instability that encompasses 80% to 85% of all CRC and adenoma[6]. These types of genetic problem always result in abnormal karyotypes such as aneuploidy, chromosome rearrangement, oncogene activation and loss of heterozygosity of tumor suppressor genes[6]. It is suggested that CIN induces carcinoma through the loss or mutation of tumor suppressor genes such as APC, TP53 and also through activation of oncogenes such as KRAS[22]. CRC caused by CIN usually have poor prognosis[23].

MSI accounts for 15% of CRC[6]. It is characterized by altered length of gene with small deletions and insertions of short repetitive deoxyribonucleic acid (DNA) sequences (microsatellite) distributed throughout the genome[6,24]. Single MSI within the whole genome may have no significant effects but accumulation of the mutations can result in frame shifts within gene coding sequences and the subsequent inactivation of the genes would give rise to the progression of tumor[6]. Mutations are frequently found in the coding mononucleotide repeats of tumor suppressors such as transforming growth factor βR2 (69%) and activin receptor type 2 (83%). Unlike the tumor caused by chromosomal instability, these kinds of tumors are diploid or near-diploid[6].

The underlying cause of MSI can be explained by two mechanisms[6]. The first mechanism is the defective mismatch repair (MMR) system, in which both alleles of a MMR gene (MLH1, MSH2, MSH6, and PMS2) are nonfunctional. This results in the loss of ability to repair DNA replication mismatches in the affected cells[22]. The germline mutation of the MMR gene can lead to HNPCC[6]. The second mechanism is the hypermethylation of promoter in MMR genes that suppress the expression of the genes like MLH1[22].

Epigenetic silencing of genes is mostly caused by DNA methylation[6]. Cancers with high degrees of methylation can be considered as CpG island methylation phenotype (CIMP) positive[25], and CIMP encompasses 35%-40% of sporadic CRC[6].

DNA methylation is involved in normal cellular control of gene expression[6]. A vast majority of methylated cytosines in the human genome are found in the CpG dinucleotide sequences. In normal cells, dense regions of CpG sequences (CpG island) are usually found in the regions close to promoters[6]. The methylation patterns of these CpG sequences are gene-specific. Aberrant CpG hypermethylation can lead to silencing of tumor-suppressor genes in carcinogenesis since the expression of the genes is repressed[25]. For example, p16, p14, MGMT and hMLH1 are commonly silenced genes in CRC patients[26]. In some cases, the presence of epigenetic silencing is overlapped with MSI[22] (Figure 2). Some sporadic CRC with microsatellite instability is caused by DNA methylation. For example, DNA methylation of MLH1 gene promoter blocks its expression and destroys the ability of MMR system[27].

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Genetic instability pathways that drive colorectal cancer development.

HNPCC: HNPCC is the most common hereditary colon cancer syndrome. It is autosomal dominant[28]. The presence of HNPCC is defined as the presence of germ-line mutation found in one of the four MMR genes, namely MLH1, MSH2, MSH6 and PMS2[29]. Bethesda guidelines can be used as a screening tool for HNPCC[30].

FAP: FAP is the most common polyposis syndrome. It is autosomal dominant and is caused by de novo germ-line mutations[31]. The presence of FAP can be diagnosed by direct sequencing of the germ-line mutations in APC gene on chromosome 5q21[32].

Since hereditary and sporadic forms of CRC are clinically different, the following discussions are focused on sporadic CRC.

CIN tumor marker: RAS family is a series of genes including HRAS, NRAS and KRAS[22]. The RAS family of oncogenes encode plasma membrane proteins at inner surface[6]. In RAS family, KRAS gene plays the most important role. KRAS gene encodes GTP (guanosine 5′-triphosphate)-binding proteins, which function as molecular switches and act as self-inactivating intracellular signal transducers for surface receptors such as epidermal growth factor receptor. The activation of RAS genes can promote cell survival and suppress apoptosis[22].

The oncogenic mutations of RAS family are believed to be an early event in CRC and they occur in 37% of CRCs and approximately 50% of adenomas > 1 cm (9% in adenomas < 1 cm)[23]. Most of KRAS mutations occur in codon 12 (70%-80%) and codon 13[23,33]. In clinical applications, KRAS mutation analysis is widely used as a prognostic and predictive biomarker of anti-EGFR monoclonal antibodies like cetuximab and panitumumab to predict the therapeutic effectiveness in CRC[34].

As well as KRAS, recent clinical studies start to focus on v-raf murine sarcoma viral oncogene homologue B1 (BRAF) and neuroblastoma-ras (NRAS). BRAF genes encode serine/threonine protein kinase that is regulated by KRA protein[35]. Mutation in BRAF gene occurs in approximately 12% of all CRC patients and it is mutually exclusive of KRAS mutation[35]. Recent studies try to investigate the clinical importance of BRAF mutation as a prognostic and predictive marker to predict resistance to anti-EGFR therapy[36]. Consideration of BRAF mutation is also recommended when KRAS mutation is not found[37]. NRAS is closely related to KRAS. NRAS mutation occurs in codon 61 and it is found in approximately 3%-5% of all CRC patients[38]. Similar to BRAF mutation, NRAS mutation is mutually exclusive of KRAS mutation[37,38].

Adenomatous polyposis coli (APC) gene plays a crucial role in the Wnt/Wingless pathway. APC gene is the most important gatekeeper of colonic epithelial cell proliferation and it is responsible for controlling the underlying oncoprotein called β-catenin. The loss of function in APC gene may lead to the transition of adenoma from normal colonic mucosa due to the up-regulation of β-catenin[22].

Germline APC mutations usually give rise to FAP. It is an inherited cancer-associated disorder in which more than 100 adenomatous polyps can be developed in mutant gene carriers. For patient with FAP, the risk of CRC by the age of 40 years is almost 100%. Somatic APC mutations are present in most sporadic colorectal adenomas and cancers. Similar to KRAS, APC mutations appear in the early stage of the progression from adenoma to carcinoma. Therefore, mutations in the APC gene are good biomarkers for identifying individuals at risk of CRC in patients’ families so as to guide the frequency of CRC surveillance and the recommendation of prophylactic surgery[22].

MSI markers: Both polymerase chain reaction (PCR) based methods and immunounhistochemistry can be used to detect MSI. Immunounhistochemistry staining detects DNA MMR system proteins such as MLH1 and MSH2. The loss in these markers is indicative of MSI. However, immunounhistochemistry staining is only best performed when the resection specimens are fixed promptly and properly since the quality of staining would be affected. Furthermore, the size of the specimens is also a concern in staining. Small specimens are not suitable for staining since the amount of tumor cells and internal staining controls are limited[6].

As PCR-based methods, currently, a 5 biomarkers panel MSI test, so-called Bethesda markers, has been developed to assess MSI status (Table 1). It consists of 2 mononucleotide loci (BAT25 and BAT26) and 3 dinucleotide loci (D2S123, D5S346, and D17S250)[39]. This panel is useful to detect about 15% of all CRC due to germline mutation in one of the mismatch repair genes (MLH1, MSH2, MSH6 and PMS2) or epigenetic slicing of MLH1. Among the five tested regions, when MSIs are present in two or more regions, the tumor is classified as MSI-High. Otherwise, the tumor is considered as MSI -Low (MSI in one region) or MSI-Stable (no MSI)[24].

The above classification system is highly valuable in prognosis and therapy since standard chemotherapy using 5-fluorouracil is not effective in treating MSI-High tumor. Instead, irinotecan-containing regimens have improved response and better prognosis for MSI-High tumor[23].

Epigenetic markers: CpG regions that are hypermethylated in CRC patients when compared to normal individuals are valuable for biomarker development. Methylations on different regions of DNA promoter have been shown to be involved in the early event of CRC development. APC methylation is one of the examples. In addition, the methylation of MLH1 associated silencing is widely used as prognostic and predictive markers for CRC[6].

Septin 9 (SEPT9) gene encodes the septin 9 protein, which is a member of GTP-binding proteins that involves in many cellular processes such as cell cycle. Disruption of SEPT9 gene could result in tumor formation[40]. In CRC development, the promoter region of SEPT9 gene was found to be hypermethylated at the early stage of CRC[41]. In the blood samples of CRC patients, the level of circulating methylated SEPT9 DNA sequences was found to be increased when compared to those of normal controls. This elevation was possibly due to apoptotic release of DNA from tumor cells into the bloodstream. The methylated SEPT9 DNA has been reported to be an effective diagnostic marker for the non-invasive detection of CRC that is independent to the ages and genders of the patients[41-44].

Unfragmented long-form DNA: In normal colonocytes, DNA is digested during apoptosis and some DNA fragments of 180-200 bps long is released into stool. However, cancerous cells in CRC show a decreased rate of apoptosis and they are not affected by the normal action of nuclear endonuclease[9]. As a result, intact unfragmented DNA sequences with 1800-2400 bp are detected in stool samples when colonocytes are shed into the lumen and then fecal stream[9,45]. The detection of unfragmented long DNA molecules in stool samples could therefore be used for CRC screening in additional to FOBT[9].

Methods for detection of gene mutation: There are a number of molecular techniques available to detect gene mutation in CRC such as allele-specific PCR, also known as amplification refractory mutation system (ARMS) that detects specific known mutated form of a gene and real-time quantitative PCR.

Both Sanger sequencing and allele-specific PCR can be used to detect the gene mutations mentioned above, such as APC and KRAS mutations. Sanger sequencing directly detects nucleotide sequences of regions of interest. Both known and novel nucleotide changes, including base substitution, insertion and deletion mutations, could be detected by this method. However, the long turnover time and high running cost make it not suitable for routine clinical uses. Allele-specific PCR is another choice for the detection of gene mutations with known positions and base changes (Figure 3). The method makes use of the allele-specific primers to selectively amplify the mutational allele sequences. Allele-specific PCR could be adopted to a real-time PCR platform in order to increase the speed and accuracy of the detection[46]. Real-time PCR employs a fluorogenic probe that is specific to the targeted DNA sequence. The probe consists of a fluorophore and a quencher attached covalently on both ends. When the probe is intact, the quencher masks any fluorescent signal emitted by the fluorophore. During PCR, the probe is first annealed to the target sequence, and then broken down by the 5’ to 3’ exonuclease activity of DNA polymerase. After the degradation of the probe, the fluorophore is released from the quencher and fluorescence is detected in real time (Figure 4). By detecting the change in fluorescent signal, the amount of target DNA molecules can be measured. With the help of real-time PCR, screening of genes mutations can be performed in batch mode and the turnover time is shorter than that of Sanger sequencing[6]. Compared to conventional PCR, real-time PCR is quantitative, fast and sensitive. As no post-PCR manipulation, such as gel or capillary electrophoresis, is required, the risk of laboratory cross-contamination can be minimized. However, real-time PCR is not suitable for studying microsatellite in which the analysis of amplification product size is required.

Open in New Tab   Full Size Figure   Download Figure

Figure 3 Principle of allele-specific polymerase chain reaction. PCR: Polymerase chain reaction.

Open in New Tab   Full Size Figure   Download Figure

Figure 4 Principle of Taqman real time polymerase chain reaction. R: Reporter; Q: Quencher.

Microsatellite markers detection: PCR amplification of specific microsatellite repeats is used for MSI detection. The presence of MSI is determined by comparing the length of nucleotide repeats in tumor cells with that of normal cells from the same patient. Normal cells adjacent to the suspicious tumor cells should be collected for comparison. The region of interest is PCR amplified with fluorescent primers and the amplification product is detected by capillary electrophoresis[24].

DNA hypermethylation detection: DNA hypermethylation can be detected in primary colorectal carcinomas using bisulfite conversion of DNA samples followed by methylation-specific PCR. Both quantitative methylation-specific PCR and pyrosequencing techniques can be employed[41,43,47,48].

MicroRNAs (miRNAs) are small non-coding RNA that is usually 19-23 nucleotides in length. Due to their small sizes, miRNAs are more stable in blood and FFPE tissues than other nucleic acids such as DNA and RNA[50]. miRNAs are involved in post-transcriptional regulation of gene expression. Therefore, they are able to function as oncogenes or tumor suppressor genes, and dysregulation of miRNA would be associated to cancers. Recent studies showed that miRNA circulate in a stable and cell-free form in bloodstream[51]. Therefore, miRNAs that are specific to CRC in blood samples may be identified for the development of non-invasive prognostic and predictive markers of the diseases[52].

MiR-135a and miR-135b play important roles in the regulation of Wnt/Wingless pathway by down-regulating APC gene expression[53]. miR-17-3p and miR-92a have been found to be elevated in plasma and CRC tissues in CRC patients, and their levels decreased after removal of the cancer tissues[54]. In the plasma of CRC patients, circulating miR-92 and miR-17 concentrations have been reported to be elevated in the preoperative samples and the concentrations were markedly reduced in the postoperative samples[51]. Those results suggested that circulating miR-92 and miR-17 are potential non-invasive diagnostic markers for CRC. Apart from the miRNAs mentioned above, miR-211 is also believed to be a potential marker for the diagnosis and prognosis of CRC[55].

Currently, a number of techniques are available for the detection of miRNA expression. PCR based methods[56], microarrays and in situ hybridization are well developed platforms for miRNA profiling. Each technique has its own advantages and disadvantages. Therefore, a suitable technique that suits the requirements of the investigation and experimental conditions should be chosen.

Quantitative reverse transcription PCR: The amount of miRNA in the specimen can be measured by real-time PCR. The first step of the measurement is to reverse transcribe miRNA into cDNA. A stem-loop primer, which enhances the miRNA reverse transcription efficiency by promoting the thermal stability of primer-miRNA complex, is usually employed. After reverse transcription, cDNA molecules are quantified by conventional real-time PCR[57] (Figure 5).

Open in New Tab   Full Size Figure   Download Figure

Figure 5 Principle of stem-loop primers. R: Reporter; Q: Quencher; PCR: Polymerase chain reaction.

Microarray: MiRNA microarray is a technique based on the hybridization between target miRNAs and an array of predesigned detection probes that are covalently immobilized onto a glass slide. The isolated miRNAs are labeled with fluorescent dye and then hybridized to a miRNA microarray. Fluorescent signal that is emitted from the labeled miRNAs at different positions on the microarray is then detected. By evaluating and analyzing the fluorescence signal data, the identities and relative quantities of miRNAs can be determined[58].

Lateral flow nucleic acid assay: Molecular diagnostic techniques developed for miRNA detection can provide highly specific and sensitive diagnostic results. However, the current existing techniques are too expensive and resource-intensive for the clinics with poor settings to perform the test[59]. In addition, well trained personnel are also a must to carry out the test and analyze the results[60]. In order to put CRC related miRNA screening into routine health check up procedure, a simple and easy-to-use detection method is desired. The emergence of the lateral flow nucleic acid assay and gold nanoparticles (Au-NPs) may help in miRNA detection[60,61]. Lateral flow nucleic acid strip can give a simple, inexpensive, fast and sensitive assay which meets the needs of point-of-care in miRNA detection[59-61].

The principle of the assay is very straightforward. First of all, the specimen is mixed with gold nanoparticles conjugated detection probe (detection probe) and biotin-bridge probe (capture probe). If target miRNA exists in the specimen, after hybridization, both the detection probe and the target miRNA bind to the complementary capture probe. The miRNA-oligonucleotide complex flows down a test strip by capillary action. The complex is eventually captured at the detection zone containing anti-avidin antibody. Nuclease is used to degrade the nonbinding capture probe and detection probe[59]. Accumulation of gold nanoparticles causes the development of a red band which can be observed by naked eyes in the detection zone. As the strength of the signal generated is proportional to the amount of the target miRNA within the specimen, to assess the result quantitatively, the strips can be scanned and imaged by a quantitative detection platform[59-61]. In order to increase the sensitivity of the detection, silver enhancement can be used. Researchers have shown that the lateral flow nucleic acid test strips and gold nanoparticles were able to detect and quantify miRNA level in a specimen as low as 1 fmol and 5 amol without and with silver enhancement, respectively[59]. This technique provides a simple, convenient and fast detection for point-of-care detection[60,61].

Novel proteomic technologies have shown promise in identification of new protein markers and profiles for early detection of CRC[62-65]. Changes in protein patterns, either due to the secretion by the tumor tissues or the non-tumor cells in tumor microenvironment, in the bloodstream can serve as potential cancer markers for CRC detection. With the fast development of proteomics, CRC specific autoantibody patterns and proteomic expression profiles can now be identified by techniques such as matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) and surface-enhanced laser desorption/ionization time-of-flight[45]. Proteomic profiling of serum from CRC patients is a promising approach to discriminate CRC patients from healthy individuals[62].

In a previous study, blood samples from individuals diagnosed with CRC and normal healthy individuals were collected. Protein expression spectra were acquired using SELDI mass spectrometry. From the results, increased levels of transferrin, alpha-1-antitrypsin and complement C3a des-arg were identified and they are treated as potential biomarkers for CRC[65]. Similar works were also carried out using either SELDI[66] or MALDI-TOF[62-64].