Research For Heterogeneity Of Multidrug Resistance In Patients With Advanced Gastric Carcinoma

Chemotherapy is important in comprehensive treatments of gastric carcinoma, especially in the patients with primary tumors resected, which is the most important therapeutic means, for this kind of cancer is insensitive to radiotherapy. But multidrug resistance (MDR) of gastric carcinoma often causes the failure of chemotherapy. MDR means that when tumor cells are resistant to a chemotherapeutic drug, they are often resistant to other drugs of different structure and mechanism. MDR contains natural MDR and acquired MDR. Natural MDR means that inherent insensitivity to chemotherapeutic drugs exist in tumor cells, so MDR will occur when a kind of chemotherapeutic drug is first used, and gastric carcinoma cells are tumor cells with obvious MDR phenotypes; acquired MDR means that tumor cells are sensitive to chemotherapeutic drugs in the beginning, but after several courses of treatment, MDR would occur, and it is common during clinical therapy for gastrointestinal tract carcinomas. MDR of gastric carcinoma causes the percentage of effectiveness is less than 50%, and it is related to low survival rate(less than 30%) of advanced gastric carcinoma. It will have great realistic meaning if the reversion of MDR for gastric carcinoma comes true.Research about MDR of tumors has been on for over 30 years. Now nearly 10 kinds of mechanism have been proven to have explicit relationship with MDR, and they are as follows:①Overexpression of MDR1/P-gp;②Overexpression of MRP;③Enhancement of exocytosis by LRP;④Reinforcement of cell's detoxification and repair effects of DNA damage;⑤The drug targets'sufficient either in quality or quantity;⑥Reinforcement of inhibitor of apoptosis proteins and reduction of apoptosis proteins;⑦Change of extracellular environment(such as pH, temperature or oxygenconcentration), hypoxia inducible factors-1alpha(HIF-1α) is research focus recently;⑧Other factors related to MDR, such as COX-2. It is generally recognized that MDR of gastric carcinoma is caused by all these factors'synergy.But with detailed study, scholars have found that many defects exist in the explanation about MDR of gastric carcinoma with these known mechanisms. The problems exist as follows:①These mechanisms can not explain all phenomena of gastric cancer MDR. The reason is most studies'objects are one/several MDR-related genes and proteins, and they often can not be used to explain all MDR phenomena in clinical studies though there often has clear conclusion in the cell lines or animal experiments.②Malignant tumors come from monoclonal proliferation, but changes in the genes or molecules often appear in the evolution of tumors. Changes (Heterogeneity) may arise in the anti-cancer drug sensitivity and expression of MDR-related factors. From a clinical point of view, most gastric carcinomas in China belong to advanced stage, and metastasis rate of lymph node reaches 60%-80%, which has been thought as independent prognostic factor, and most patients died of gastric cancer recurrence / or metastasis after primary tumors were resected. Current chemotherapy regimens often are experiential programs, and the efficacy is poor. Some reports have shown that chemosensitivity test in vitro of tumor cells can be used as evidence in the selection of chemotherapeutics, but these results cannot reflect condition of multi-channel and multi-factors in vivo, also they cannot be on behalf of the results of lymph node metastases (LNMs). MDR in LNMs of gastric carcinoma has been rarely reported in the literature both at home and abroad.Our study of these series is based on the problems above, and we prepare to conduct the following study of gastric cancer MDR. Our study is divided into two series as follows:①Series 1 is divided into 4 four parts, the relationship between MDR related factors and chemosensitivity in vitro has been investigated from the most established ways that lead to MDR of gastric carcinoma; characters of MDR in LNMs are also studied, and heterogeneity is analyzed compared with results of PT.②Series 2 is to find new key factor of MDR. Advanced techniques of proteomics are used to screen and identificate differentially expressed proteins in different gastric carcinoma cell lines. This study may be helpful in mechanism and reversal in MDR of gastric carcinoma.Protein is the direct executor of the life's structure and activities, so it's more important to investigate protein than gene. Proteome means all the proteins encoded by genome and proteomics take all proteome encoded by genome as object, then it analysis dynamic changes of intracellular protein composition and activities of law as a whole. Research of proteomics depends on three techniques: technique of protein components'separation, identification technology of protein components and technique of function prediction. Proteomics brings new mode of thinking in the research of MDR about malignant tumors. Proteomics studies mechanism of MDR about malignant tumors from the overall level of tissue or cell protein. In the methodological, it's unnecessary to determine the definite object of study, and also it's unnecessary to prepare antibody in advance. This greatly broadens the scope of the study, and we can be more accurate analysis of life phenomena.The main contents of this article are as follows: Part one: Chemosensitivity of regional lymph node metastases compared to the primary tumors with SRB assay in vitro for patients with advanced gastric carcinomaObjective: To investigate the different chemosensitivity of lymph node metastases cultures compared to the primary tumor in gastric carcinoma gastric carcinoma by sulphorhodamine B (SRB) assay.Methods: In vitro SRB assay was performed on 56 paired surgical specimens of primary tumors (PT) and lymph node metastases (LNMs) from 56 gastric carcinoma patients. The results were then compared with response to 11 antitumor drugs(5-FU, VP-16, HCPT, PTX, L-OHP, CDDP, eADM, THP, MMC, VCR, MTX). Data was dealt with t test and Spearman correlation analysis.Results: 1.There was significantly different chemosensitivity of individuals to antitumor drugs between PT and LNMs in 7/11 drugs(all P<0.05). The inhibition rates of tumor cells in LNMs for HCPT, L-OHP and VCR were lower than those in PT(all P<0.05), whereas the inhibition rates of VP-16, eADM, THP, MMC were higher in LNMs than those in PT(all P<0.05). No significantly difference of inhibition ratio between PT and LNMs was found in 5-FU,PTX,CDDP,MTX(all P>0.05). 2.There was a significantly positive correlation between the inhibition rates of PT and LNMs for 5-FU,VP-16, HCPT, L-OHP,PTX,THP,MTX (r=0.3735~0.5177, all P<0.05).Summary: 1. SRB assay has many advantages such as high sensitivity, pin-point accuracy, more simple operation, lower cost, etc. So it's worthy of promotion in clinic. 2. All 11 drugs have low inhibition rates for gastric carcinoma cells , and it shows that natural MDR exists in gastric carcinoma cells, so it's unrealistic to control this cancer with one drug. 3. Heterogeneity of chemosensitivity to most antitumor drugs is shown in metastatic lymph nodes in gastric cancers. The chemosensitivity of gastric carcinoma primary tumor in the drugs response assay in votro may not give a good indication of the chemosensitivity in metastatic lesions. Chemotherapy after operation should be carried on according to characteristic of LNMs in patients with advanced gastric carcinoma.Part two: Relationship between expression of P-gp, GST-π, TopoⅡαand chemosensitivities in primary tumors(PT) and regional lymph node metastases(LNMs) of gastric carcinomaObjects: The most studied classical pathways of MDR include drug pump of MDR1/P-gp, detoxification of GST-πand drug targets regulation of TopoⅡα. These pathways are most viewed. In this part, expression of P-gp, GST-π, TopoⅡand chemosensitivities were detected in PT and LNMs, their relationship was analyzed, and the significance that may exist was discussed.Methods: In vitro SRB assay was performed on 56 paired fresh surgical specimens of PT and LNMs from 56 gastric carcinoma patients. Expression of P-gp, GST-π, TopoⅡwas tested by immunohistochemical staining. Measurement data were analyzed with t test, ranked data with Wilcoxon rank test, and Spearman correlation analysis was also used to calculate correlation coefficient.Results: 1.The results of chemosensitivity in vitro in tumor cells of PT and LNMs were seen in Part one. 2. The expression of P-gp and GST-πwas higher in LNMs than in PT(both P<0.01), and expression of TopoⅡαwas higher in PT(P<0.01). There was positive correlativity of P-gp between PT and LNMs(r=0.3303, P<0.05), and no significant correlativity was found about GST-πand TopoⅡαbetween PT and LNMs(both P>0.05). 3. In PT the inhibition rates for PTX, HCPT, CDDP in P-gp strong expression group were lower than those in weak group (all P<0.05); The inhibition rate to CDDP was significantly lower for the GST-πstrong expression group in PT (P<0.05); and for strong expression of TopoⅡα, the inhibition rates for VCR, L-OHP were higher(both P<0.05). In LNMs, there were lower inhibition rates for VP-16, MTX in P-gp strong expression group (both P<0.05), and the inhibition rate to PTX was significantly lower for the TopoⅡαweak expression group in LNMs(P<0.05); no significant change was found between inhibition rates for these drugs and expression of GST-π(all P>0.05).4. The inhibition rates to HCPT, PTX, CDDP were negatively correlated with the expression of P-gp in PT (r=-0.2856, r=-0.2967,P=0.0264; all P<0.05), and negative correlation was found between the inhibition rates to VP-16, MTX and the expression of P-gp in LNMs(r=-0.2753, r=-0.2665; both P<0.05). The expression of GST-πwas negatively correlated with the inhibition rates to 5-Fu, VP-16 in PT( r=-0.3475, r=-0.2913; both P<0.05), and negatively correlated with HCPT, VCR(r=-0.3703, r=-0.3218;both P<0.05). The expression of TopoⅡαwas positively correlated with inhibition rates to L-OHP in PT(r=0.3243, P<0.05), and positive correlation was found between the expression of TopoⅡαand inhibition rates to PTX in LNMs(r=0.4085, P<0.05).Summary: 1. There was higher positive expression rate of P-gp, GST-π, TopoⅡαbetween PT and LNMs of gastric carcinoma, and expression of these proteins showed heterogeneity between PT and LNMs. It's of great significance to test expression of P-gp, GST-π, TopoⅡαin LNMs, and it can reflect the MDR status in patients after PT resected. 2. There was heterogeneity in the relationship about expression of P-gp, GST-π, TopoⅡαand chemosensitivity in 11 chemotherapeutics between PT and LNMs. The results showed that these factors all participate in MDR of gastric carcinoma, and there existed relationship between expression of them and chemosensitivity in part of chemotherapeutics, but expression of them could not reflect chemosensitivity of all drugs. So it was inaccurate to predict MDR of LNMs with results of PT. 3. P-gp, GST-π, TopoⅡαcould not explain all phenomena of MDR, so it's of great significance to search new MDR related factors to explore the mechanism of MDR of gastric carcinoma.Part three: Relationship between expression of apoptosis-related proteins (p53, Survivin, Bcl-2, Bax) and chemosensitivities in primary tumors (PT) and regional lymph node metastases (LNMs) of gastric carcinomaObjective: Many recent researches have shown that regulatory mechanisms of apoptosis of tumor cells play an important role in MDR of gastric carcinoma.So in this part, expression of apoptosis-related proteins (p53, Survivin, Bcl-2, Bax) and chemosensitivities in vitro were detected in PT and LNMs, their relationship was analyzed, and the significance that may exist were discussed.Methods: In vitro SRB assay was performed on 56 paired fresh surgical specimens of PT and LNMs from 56 gastric carcinoma patients. Expression of p53, Survivin, Bcl-2, Bax was tested by immunohistochemical staining. Measurement data were analyzed with t test, ranked data with Wilcoxon rank test, and Spearman correlation analysis was also used, and correlation coefficient was calculated.Results: 1.The results of chemosensitivity in vitro in tumor cells of PT and LNMs were seen in Part one. 2. The expression of Bcl-2, Bax was higher in LNMs than in PT(both P<0.01), and there was no significant difference in expression of p53, Survivin between PT and LNMs(both P>0.05). There was positive correlativity of p53, Bcl-2, Bax between PT and LNMs(r=0.7255, r=0.4262, r=0.2901; P<0.05). 3. In PT the inhibition rates for PTX, CDDP in p53 strong expression group were lower than those in weak group (both P<0.05); the inhibition rates to VCR, PTX, eADM were significantly lower for the Survivin strong expression group in PT (all P<0.05), but for L-OHP, the inhibition rate was higher when the tumor with stronger expression of Survivin(P<0.05); and for strong expression of Bcl-2, the inhibition rates for 5-FU, PTX, eADM, HCPT were lower(all P<0.05); for strong expression of Bax, the inhibition rates for 5-FU, eADM, HCPT were higher(all P<0.05). In LNMs, there was lower inhibition rate for PTX in p53 strong expression group (P<0.05), the inhibition rate to MTX was significantly lower for the Survivin strong expression group in LNMs(P<0.01); and for strong expression of Bcl-2, the inhibition rates for 5-FU, VCR, VP-16, PTX were lower(all P<0.05); no significant change was found between inhibition rates for these drugs and expression of Bax(all P>0.05).4. There was negatively correlation between inhibition rate to CDDP and the expression of p53 in PT(P<0.01) , and no significant relationship was found between expression of p53 and inhibition rates for these 11 drugs(all P>0.05). The inhibition rates to 5-FU, PTX, eADM were negatively correlated with the expression of Survivin in PT (r=-0.4805, r=-0.2979, r=-0.2861; all P<0.05), and there was positive correlation was found between inhibition rate to L-OHP and expression of Survivin(r=0.3329, P<0.05); negative correlation was found between the inhibition rates to MTX and the expression of Survivin in LNMs(r=-0.3611; P<0.01). The expression of Bcl-2 was negatively correlated with the inhibition rates to 5-FU, VP-16, HCPT, PTX, eADM in PT(r=-0.3722, r=-0.3032, r=-0.5035, r=-0.3973, r=-0.2712;all P<0.05), and negatively correlated with 5-FU, VP-16, PTX, VCR(r=-0.2737, r=-0.6238, r=-0.3918, r=-0.3469;all P<0.05)in LNMs. The expression of Bax was positively correlated with inhibition rates to 5-FU, HCPT, eADM in PT(r=0.4377, r=0.3666, r=0.3517;all P<0.05), and no significant correlation was found between the expression of Bax and inhibition rates to these drugs in LNMs(all P>0.05). Summary: 1. There was heterogeneity in expression of apoptosis-related proteins (p53, Survivin, Bcl-2, Bax) between PT and LNMs. The results showed that it's valuable to test these factors in LNMs to reflect situation of MDR in patients with PT resected. 2. There was heterogeneity in the relationship between expression of p53, Survivin, Bcl-2, Bax and chemosensitivity in 11 chemotherapeutics between PT and LNMs. The results showed that these factors all participate in MDR of gastric carcinoma, and there existed relationship between expression of them and chemosensitivity in part of chemotherapeutics, but expression of them could not reflect chemosensitivity of all drugs. So it was inaccurate to predict MDR of LNMs with results of PT. 3. Results showed that p53, Survivin, Bcl-2, Bax all participate in MDR, but none of them could explain all phenomena of MDR, so it's of great significance to search new key MDR related factors to explore the mechanism of MDR of gastric carcinoma.Part four: Relationship between expressions of cyclooxygenase-2(COX-2), MDR related factors and chemosensitivities in vitro in primary tumors (PT) and regional lymph node metastases (LNMs) of gastric carcinomaObjective: Many researches in recent years have shown that cyclooxygenase-2(COX-2) has close relationship with MDR of gastric carcinoma, and it can participate in MDR gastric carcinoma by regulating expression of other of MDR related factors. So in this part, expression of COX-2, some other MDR related factors and chemosensitivities in vitro were detected in PT and LNMs, their relationship was analyzed, and the significance that may exist was investigated.Methods: In vitro SRB assay was performed on 56 paired fresh surgical specimens of PT and LNMs from 56 gastric carcinoma patients. Expression of COX-2, some other MDR related factors were tested by immunohistochemical staining. Statistical analysis was the same as Part three.Results: 1.The results of chemosensitivity in vitro in tumor cells of PT and LNMs were seen in Part one. 2. The expression of COX-2 was higher in LNMs than in PT(P<0.01), and there was positive correlativity of COX-2 between PT and LNMs(r=0.3511, P<0.01). 3. In PT, there was positive correlation between the expression of COX-2 and other MDR related factors as following: GST-π, Survivin, Bcl-2(r=0.2741,r=0.7024,r=0.5925;all P <0.05), and negative correlation was found between expression of COX-2 and TopoⅡα, Bax(r=-0.3488, r=-0.3292,both P <0.05). In LNMs, positive correlation was found between expression of COX-2 and P-gp, Survivin, Bcl-2(r=0.3490,r=0.2836,r=0.5368;all P <0.05). 4.In PT the inhibition rates for 5-FU, VCR, PTX, eADM, HCPT in COX-2 strong expression group were lower than those in weak group (all P<0.01). In LNMs, there were lower inhibition rates for 5-FU, VCR, VP-16, MTX in COX-2 strong expression group (all P<0.05).5. The inhibition rates to 5-FU, HCPT, PTX, eADM, THP, VCR were negatively correlated with the expression of COX-2 in PT (r=-0.5385,r=-0.3077,r=-0.3850,r=-0.2991,r=-0.3466,r=-0.4158;all P <0.05), and negatively correlated with 5-FU, VP-16, PTX, VCR, MTX(r=-0.4120, r=-0.4421, r=-0.3300,r=-0.3597, r=-0.2836;all P <0.05)in LNMs.Summary: 1. There was heterogeneity in expression of COX-2 between PT and LNMs, and positive correlativity was found between expression of COX-2 in PT and in LNMs, which showed that COX-2 played an important role in metastasis of gastric carcinoma. 2. There was heterogeneity in the relationship about COX-2 and P-gp, GST-π, TopoⅡαbetween PT and LNMs. 3. There was heterogeneity in the relationship of COX-2 and p53, Survivin, Bcl-2, Bax between PT and LNMs. 3. Results showed that COX-2 participates in MDR of gastric carcinoma, but it could not explain all phenomena of MDR, and it is uncertain about the effect of inhibitor of COX-2 in the clinical therapy. So it's of great significance to search new key MDR related factors to explore the mechanism of MDR of gastric carcinoma.Part five: Identification of differentiation-related proteins in gastric carcinoma cell lines by comparative proteomics and their relationship with MDR of gastric carcinomaObjective: Researches have shown that MDR of digestive tract was in related to differentiation of tumor cells. So in this part, differentiation of gastric cancer cell lines were used as object to investigate differentiation-related proteins in human gastric carcinoma cell lines by comparative proteomics, and then their relationship with MDR of gastric carcinoma was studied.Methods: 1 The holoproteins of human gastric carcinoma cell lines MKN28(well differentiated), SGC7901(moderately differentiated) and BGC823(poorly differentiated) were measured by two-dimensional gel electrophoresis and matrix assisted laser desorption/ionization time-of-flight mass spectrometry(MALDI-TOF-MS).Some proteins obtained through proteomics were tested by Western blot in the cell strain and tissues of gastric carcinoma.Results: There were different proteins in human gastric carcinoma cell lines. All 14 different protein spots were found in 3 gastric carcinoma cell lines, and 8 spots were identified by MALDI-TOF-MS, and these proteins were probable thioredoxin peroxidase, glyceraldehyde-3-phosphate dehydrogenase (GAPD),β-tubulin polypeptide, hypothetical protein, zinc finger protein (ZNF)139, protein-tyrosine kinase, calreticulin precursor, tropomyosin. These proteins have relationship with the biological behavior of gastric carcinoma, such as signal transduction, cellular homeostasis, glycolysis, antioxidation action, multidrug resistance (MDR), etc. Result of Western blot about protein expression was in consonance with situation of proteomics.Summary: 1. Proteins spots got from 3 gastric cells were all 1000 more or less, and 8 spots of 14 were identificated, part of them verificated by Western blot, which showed it was credible of the results. 2. These 8 proteins were involved in many biological behaviors of gastric carcinoma, such as signal transduction, cellular homeostasis, glycolysis, antioxidation action, multidrug resistance (MDR), etc. The results showed that intervention for these proteins could reverse malignant some biological behaviors of gastric carcinoma. Part six: Identification of gastric carcinoma cell line SGC7901 and tumor MDR cell strain SGC7901/VCR by comparative proteomicsObjective: To obtain new proteins directly related to the MDR of gastric carcinoma, comparative proteomics was used in this part, and gastric carcinoma cell line SGC7901 and tumor MDR cell strain SGC7901/VCR were used as objects to identificate new differently expressed proteins.Methods: Methods were the same as Part five.Results: There were different proteins in these 2 gastric carcinoma cell lines. All 9 different protein spots were found, and 7 spots were identified by MALDI-TOF-MS. These proteins were 60S ribosomal protein L23, voltage-dependent anion-selective channel protein (VDAC1), zinc finger protein (ZNF)394, keratin, type I cytoskeletal 9(KⅠC9), RNA 3'-terminal phosphate cyclase(RTC1), zinc finger matrin-type protein 2, Sideroflexin-1. These proteins have relationship with the biological behavior of gastric carcinoma, such as multidrug resistance (MDR), cellular signal transduction, maintenance of cellular homeostasis, proliferation and apoptosis, maturation of cells, etc. Result of Western blot about protein expression was in consonance with situation of proteomics.Summary: 1. Proteins spots got from 3 gastric cells were all over 1000, and 7 spots of 9 were identificated, parts of which were certificated by Western blot assay, which showed it was credible of the results, and these proteins were directly related to the MDR of gastric carcinoma. 2. These 7 proteins were involved in many biological behaviors of gastric carcinoma, such as multidrug resistance (MDR), cellular signal transduction, maintenance of cellular homeostasis, proliferation and apoptosis, maturation of cells, etc. The results showed that intervention for these proteins could reverse MDR of gastric carcinoma, improve the effect of chemotherapy, and improve the prognosis. Conclusions:1. There was heterogeneity in chemosensitivity in vitro related between PT and LNMs, so it's inaccurate to design chemotherapy for patients with PT resected according to results of chemosensitivity test for PT, it should be designed to aim at MDR features of LNMs.2. There was heterogeneity in expression of MDR related factors(P-gp, GST-π, TopoⅡα, p53, Survivin, Bcl-2, Bax) between PT and LNMs. Heterogeneity was also found in the relationship about expression of MDR related factors and chemosensitivity in chemotherapeutics between PT and LNMs. The results showed that these factors all participate in MDR of gastric carcinoma, and there existed relationship between expression of them and chemosensitivity in part of chemotherapeutics, but expression of them could not reflect chemosensitivity of all drugs. It's of great significance to search new key MDR related factors to explore the mechanism of MDR of gastric carcinoma.3. Results showed that COX-2 participates in MDR of gastric carcinoma, and there was heterogeneity in expression of COX-2 between PT and LNMs. Relationship between COX-2 and MDR related factors was different from that in LNMs, and heterogeneity was also found in the relationship about expression of COX-2 and chemosensitivity in chemotherapeutics between PT and LNMs.4. 8 spots of 14 differentiation-related proteins were identificated, which were probable thioredoxin peroxidase, glyceraldehyde-3-phosphate dehydrogenase (GAPDH),β-tubulin polypeptide, hypothetical protein, zinc finger protein (ZNF)139, protein-tyrosine kinase, calreticulin precursor, tropomyosin; part of them were certificated by Western blot. In these proteins,β-tubulin polypeptide and protein-tyrosine kinase were in related to MDR of gastric carcinoma.5. 7 proteins directly related to the MDR of gastric carcinoma were identificated by comparative proteomics in gastric carcinoma cell line SGC7901 and tumor MDR cell strain SGC7901/VCR, which were 60S ribosomal protein L23, voltage-dependent anion-selective channel protein (VDAC1), zinc finger protein (ZNF)394, keratin, type I cytoskeletal 9(KⅠC9), RNA 3'-terminal phosphate cyclase(RTC1), zinc finger matrin-type protein 2, Sideroflexin-1; part of them were certificated by Western blot assay.