br Corresponding author br closely associated with the progn
closely associated with the prognosis of patients with advanced GC [8–10]. Therefore, identification of the phenotype of GC differentiation will facilitate the implementation of effective chemotherapy regimens and accurate predict patient outcomes. Currently, pathologists primarily judge histological grades, such as differentiation status, based upon changes in tissue/cell morphology after haematoxylin-eosin staining. This judgement depends largely on the subjective impression and discretion of the pathologist, making it difficult to achieve accurate and sensitive results . GC develops through the accumulation of genetic and epigenetic alterations, and molecular alterations occur even before the development of pre-cancerous lesions [12–14]; therefore, molecular marker-based histolo-gical detection is more conducive to evaluating cancer progression, including tumour differentiation status, with higher accuracy and sen-sitivity. Mirza et al. demonstrated that if the colon epithelial protein was detected in a patient's non-cancerous sample, this patient was more likely to develop GC compared with patients for whom the protein was not detected (93% vs 35%, P < 0.0001) . CD44 is more frequently
and highly expressed in moderately differentiated GC, whereas alde-hyde dehydrogenase (ALDH) is correlated with poorly differentiated GC . However, at present, there is still a lack of sensitive and specific molecular markers available for the differentiation status of GC.
Recently, an iterative process known as “Cell-Systematic Evolution of Ligands through Exponential Enrichment (Cell-SELEX)” has been established and can be employed for the targeting of a specific cell type without any prior knowledge of the specific target, thereby enabling the identification of multiple ligands and the ability to discriminate among even closely related cell phenotypes [16,17]. The ligands, which are called aptamers, that are selected using the Cell-SELEX technique are short (20–100 bases) single-stranded DNA (ssDNA) or RNA molecules that bind with high affinity and specificity to their target substances through conformational complementarities. With the application of an additional subtractive step, termed as subtractive selection, the selected aptamers can specifically recognize the phenotypic variations between target PepstatinA and subtractive cells even if the target is unknown, allowing the aptamers to correctly distinguish tumour cells from healthy cells and to discriminate between cell types with different properties (i.e., tumourigenesis and metastatic potential) [18–20]. Shangguan et al. applied this approach to identify aptamers that, by specifically binding T-cell acute lymphocytic leukaemia cells, could distinguish these cells from B-cell lymphoma cells, as well as a mixture of normal human bone marrow aspirates . Our group performed the differential selection strategy using metastatic colorectal cancer (CRC) cells and non-meta-static cells, resulting in aptamers that specifically bind to metastatic CRC cells .
Aptamers are also referred to as “chemical antibodies”. However, compared to monoclonal antibodies, aptamers possess a number of important advantages, including high affinity and specificity, efficient and cost-effective chemical synthesis, easy and controllable modifica-tion, rapid tissue penetration, and low levels of toxicity and im-munogenicity, making aptamers promising molecular probes for bio-logical applications, particularly in the fields of imaging pathology and biomarker identification [21–23]. Zamay et al. described a method for staining tumour cells and blood vessels in lung cancer tissue using DNA aptamers as imaging probes for intraoperative tumour visualization . Bing et al. used the aptamer Sgc-3b to identify the target protein selectin L, a biomarker found on naive T cells that can distinguish central memory from effector memory T cells . In summary, com-pared to other biomarker selection technologies, the aptamers selected by Cell-SELEX not only serve as probes towards cells with given phe-notypes but also facilitate biomarker discovery.
To date, the selection of certain GC-specific aptamers has been reported using subtractive Cell-SELEX [24–26]. Zhang et al. used the GC cell line HGC-27 as target cells and GES-1 as control cells for generating aptamer AGC03, which could specifically bind to GC cells as well as paraffin-embedded cancer tissue sections but not to adjacent non-cancerous tissues . However, no information was provided regarding the ability of this aptamer to resolve different types of GC, particularly poorly differentiated GC with malignant potential. In this paper, for the first time, based on the degree of differentiation be-tween the two human GC cell lines [27,28], we used the poorly dif-ferentiated GC cell line BGC-823 as the target cells and the moderately differentiated GC cell line SGC-7901 as the negative cells to perform subtractive Cell-SELEX. Two aptamers, PDGC21and PDGC45, were generated that specifically bind to BGC-823 cells. After being trun-cated, aptamer PDGC21-T exhibited a better affinity compared with the full length aptamer PDGC21. Moreover, aptamer PDGC21-T was conjugated to quantum dots (QD605) as a specific molecular probe for the targeted imaging of cancer cells in mixed culture cells and tissues from patients with GC, especially in poorly differentiated GC tissues, indicating that aptamer PDGC21-T holds great potential for use as a molecular probe for detecting and targeting poorly differentiated GC, which can, in turn, direct drug treatment and predict outcomes in clinical trials.