The Discovery And Identification Of Two Novel Biomarkers On The Surface Of Gastric Cancer Cells

Gastric cancer (GC) is a common neoplasm in digestive system, which is leadingthe cause of cancer-related deaths worldwide. With the progression of earlydiagnostic techniques and therapeutic strategies, the overall survival rate of GCpatients has been improved in recent years. The outcome survival rate of theadvanced GC still remains poor while the5-year overall survival rate is diminishing.In post-genomic era, the discovery and identification of novel GC biomarkers isrequired for more sensitive, specific and functional targets, which could be easilyused for diagnosis and therapeutic purposes. The generaton and the development ofmonoclonal antibody would be played a key role in the basic and clinical researchesof GC.The development of proteomics has been set the feasible approches to identifythe GC biomarkers and the associated molecules although many known biomarkers ofGC are non-specific and limited. Therefore, the HTS (high throughput screening) andidentification of specific molecules, especially the membrane bound proteins fromdifferent cancer cells, would provide the new hopes besed on monoclonal antibodytechnology to diagnosis and treat the GC patients.In this study, we have used an antibody-based approach to identify antigen withthe counter screening strategy. The aim of our study was to generate monoclonalantibodies (mAbs) against the native antigens or receptors as the cancer biomarkerson the surface of gastric cancer cells through our unique “one step” colony pickingand HTS screening methods. It is difficult to prepare mAbs against the nativeantigens on cell surface if still using the current screening approaches. Therefore, we have developed a streamlined strategy to make mAbs binding to native antigens oncell surface, which we used live cell immunization(so-called “shot gun” manner) andlive cell screening FACS-HTS (fluorescence activating cell sorter-highthrough putscreening) strategy with the unique hybridoma technology. The identifed and selectedmAbs were implemented in the Immunohistochemstry, ELISA, FACS analysis and ahuman proteome microarray etc. These biomarkers were verified at least twice by thetwo mAbs.The four sections of our study as follows:Ⅰ. The specific mAbs against native antigens on cell-surface of gastric cancercells were generated and screened by FACS-HTSMETHODS: Gastric cancer cell lines (SGC7901, BGC823, MKN28, MKN45)were cultured and mixed together to be served as an immunogen in about1×107cellsper mouse were sc.(subcutaneous) and ip.(intraperitoneal) injections into6-8weeksold BALB/C mice. The splenocytes from three times immunized mouse were fusedwith SP2/0myeloma cells and inoculated in96-wells cell culture plates with a largescale manner. The immunoreactivity hybridoma from the culture supernatants of HATconditional media were used to screen with live gastric cancer cells, as well as thenormal control cells (peripheral blood mononuclear cells from healthy individuals) inFACS-HTS system. We selected many positive hybrids according to ELISA-HTSprinciple and double confirm the signals, positive and stable hybridoma clones werevalidated and picked and then the immunoglobulin isotyping were performed. Thepurification of mAbs by Protein A/G-Sepharose chromatography were conducted.RESULTS: Two mAbs against native antigens on cell-surface of gastric cancercells, termed as MS38-2.1and MS40-3.1, were generated via hybridoma technologyin combination with viable cell-based FACS-HTS. Both of the two mAbs were IgG1/ κ.CONCLUSIONS: The series data indicated that our strategy for the generationand validation of anti-GC mAbs is working well. The “shot gun” mannerimmunization and the combination screening methods that the supernatants ofhybridoma colonies were for both ELISA-HTS and the live cells based FACS-HTSare the pivotal high throughput approaches in a robust platform for the mAbgenerations that would be specifically for the native antigens on cell surface of gastriccancer and other solid tumors.Ⅱ. The validation of mAbs specificity against native antigen on the surface ofgastric cancer cellsMETHODS: Having38cell lines including15gastric cancer cell lines,colorectal cancer, liver cancer, breast cancer, gallbladder cancer, acute monocyticleukemia and GES-1, EA.hy926cells were used to analyse the binding profiles ofMS38-2.1and MS40-3.1through indirect immunofluorescence (IF) staining andFACS assay. The12fresh tumor tissues and the adjacent normal mucosa tissues werefreshly obtained from the gastric cancer patients who were undertaken the surgeryresection. A proportion of12fresh gastric tissues were separated into single-cellsuspension by a cocktail enzymes digestion that would be immediately for FACSanalysis with indirect IF staining by MS38-2.1and MS40-3.1mAb. Anotherproportion of12fresh gastric tissues were used for immunohistochemical (IHC)analysis with formalin-fixed and paraffin-embedded (FFPE) to identify theimmunoreactivity of two mAbs in tissues.RESULTS: The MS38-2.1and MS40-3.1mAbs demonstrated bothimmunofluorescence and immunohistochemistry grades from the assays. The twopurified mAbs specifically recognized the native antigens on cell surface of gastric cancer and other malignancies, and the MS38-2.1could interestingly haveimmunoreactivity with EA.hy926cells (human umbilical vein endothelial cells). Thetwo mAbs have no cross reactions to the normal control cells including humanperipheral blood mononuclear cells (PBMC) and normal primary gastric cells. Thesignificant IHC staining on the gastric cancer cells were observed when detected bythe two mAbs and strong staining of small and medium blood vessels were observedwhen detected with MS38-2.1. FACS analysis demonstrated that there weresignificant differences between fresh prepared gastric cancer cells and correspondingadjacent normal gastric cells when detected by MS38-2.1and MS40-3.1mAbsrespectively (P <0.05).CONCLUSIONS: The target antigen recognized by MS38-2.1was expressed ingastric cancer, colorectal cancer, liver cancer, breast cancer, gallbladder cancer, acutemonocytic leukemia, GES-1and L-02cells, and also in vascular endothelial cells ofsmall and medium blood vessels.Ⅲ. The correlation and clinical significance between the expression of gastriccancer cell-surfaced native antigens defined by specific mAbs andclinicopathological characteristics of gastric cancerMETHODS:90FFPE gastric cancer tissues and corresponding adjacent normalmucosa tissues were selected to the applications of the immunoreactivity ofMS38-2.1and MS40-3.1mAbs in IHC. The relationship among the staining intensityof two mAbs, clinicopathologic characteristics and prognosis were evaluated.RESULTS:28in90cases were positive (31.11%) on both gastric cancer tissuesand their small blood vessels, and8in90cases were positive (8.89%) on tumoradjacent tissues and their small blood vessels by MS38-2.1mAb (P<0.05).42in90cases were positive (46.67%) on gastric cancer tissues and only3out of90cases arepositive (3.33%) on tumor adjacent tissues by MS40-3.1mAb (P<0.05). The IHC staining of both MS38-2.1and MS40-3.1mAbs showed the specificifity on thetargets that are related to degrees of tumor invasion, lymph nodes metastasis andTNM stage, but not correlated to age, gender, the size and location of tumor andtumor differentiation. The survival analysis demonstrated that there was a significantcorrelation between short survival and high expression of target antigens thatMS38-2.1and MS40-3.1specifically binding to (P＜0.0001).CONCLUSIONS: The biomarkers that were recognized by MS38-2.1andMS40-3.1mAbs were both specifically expressed in gastric cancer tissues and relatedto tumor invasion, metastasis, clinical staging and prognosis, which could bepotentially applied to the prognosis assessment and diagnosis of gastric cancer.Ⅳ. The identification of the native antigens on cell surface of gastric cancerspecifically recognized by mAbsMETHODS: Gastric cancer cell lines (BGC823, MKN28, SGC7901, MKN45),human gastric epithelial cells (GES-1), colon adenocarcinoma cells (SW1116),breast cancer cells (MCF-7) and hepatocellular adenocarcinoma cells (Hep3B) wereall cultured and treated with mammalian cell membrane protein lysis buffer. TheWestern blot method was utilized to determine the expressed proteins and theirmolecular weights of these cell lines, which MS38-2.1and MS40-3.1were used forspecific binding. The lysates of normal gastric tissues were served as the control. The16,368unique full-length human proteins in proteome microarray were used toanalyze the specificity of the two mAbs and to identify the target candidates or theprotein ID of these two mAbs. The immunoreactivity of the potential target antigenswith the two mAbs was validated with ELISA and Western blot.RESULTS: The results of Western blot suggested the biomarker candidates thatMS38-2.1specifically binding to was highly expressed on cell surface, includinggastric cancer cell lines (BGC823, SGC7901, MKN45) and human gastric epithelial cell line GES-1, but was lowly expressed on gastric cancer cell line MKN28. Inaddition, the biomarker candidate was expressed on colon adenocarcinoma cell lineSW1116, breast cancer cell line MCF-7and hepatocellular adenocarcinoma cell lineHep3B. However, the lysates of normal gastric tissues not comprised the biomarkercandidates which were detected by MS38-2.1. The approximate molecular weight ofthe biomarker candidate was128kDa.As for MS40-3.1, the Western blot demonstrated that the biomarker candidatethat the mAb could specifically bind to was highly expressed on the cancer cellsurface, including gastric cancer cell lines BGC823, SGC7901, MKN28, MKN45andhuman gastric epithelial cells GES-1. The biomarker candidate was also expressed oncolon adenocarcinoma cells SW1116and hepatocellular adenocarcinoma cells Hep3B.However, the lysates of normal gastric tissues not comprised the biomarkercandidates which was detected with MS40-3.1. The approximate molecular weight ofthe biomarker candidate was60kDa.The results of human proteome microarray showed that the both generated mAbsare monospecific antibodies. The biomarker candidate that MS38-2.1specificallybinding to is a small proline-rich protein3(SPRR3). The biomarker candidate thatMS40-3.1specifically binding to is an ornithine decarboxylase1(ODC-1). TheELISA and Western blot assay showed that the favourable immunoreactivitiesbetween human recombinant proteins SPRR3, ODC-1and MS38-2.1, MS40-3.1respectively.CONCLUSIONS: Both MS38-2.1and MS40-3.1are monospecific antibodies.The biomarker candidates that MS38-2.1, MS40-3.1specifically binding to are somemolecules similar to SPRR3, ODC-1respectively, or at least molecules share thecommon epitopes of SPRR3and ODC-1. According to current results, The twobiomarker candidates might be novel biomarkers expressed on gastric cancer cell surface.In general, our study has been developed a streamlined strategy for the screeningof the monospecific mAbs against gastric cancer cell surface native conformationalantigens with the combination of viable cell immunization, hybridoma techniques andthe live cell based FACS-HTS assays. The two potential biomarkers on gastric cancercell surface were preliminarily identified and might be applied for diagnosis andprognosis assessment of gastric cancer. The novel approaches developed in our studyand the current results will be of great importance to the research work of molecularbiomarkers of various human malignancies.