| Background:Along with the rapidly developing on the theory of molecular biology and the improving on the experimental technology, the exploration of biomarkers had turned into a hot researching area. The measurement and analysis on the key biological macromolecules were helpful for the diagnosis on diseases in the early stage and improved their treatment and prognosis. The discovery of biomarkers included finding or identification, validation, and the confirmation from clinical center these three steps. On the stage of finding and identifying the candidate markers, unite multiple omics was a critical technique to select the differentially expressed genes or proteins. Currently, this technique is a popular researching topic on analyzing and selecting the differentially expressed factor on diseases within the level of gene and the level of protein. On the stage of validating the serum potential protein biomarkers, at least two different validating techniques were needed, and the results of them had to match each other. The verification of serum potential protein biomarkers for some infectious diseases were difficult to complete the experiment on animals because of their particularity and imitated conditions; however, multiple reaction monitoring (MRM) was able to solve this kind of problem. The benefits of operating MRM were not needed for the preparation of antibodies, low cost, high throughput, and testing multiple proteins at a same time. Therefore, it became highly advanced on the stage of validating the serum potential protein biomarker; also, MRM was considered as a gold standard of using mass spectrometry to verify the markers. In summary, our research used transcriptomics, quantitative proteomics, and MRM to select and verify the differentially expressed proteins in serum of liver cancer and smear-negative tuberculosis. Furthermore, our research discussed the possibility of serum amyloid A1 (SAA1) and ficolin 3 (FCN3) that functioned as the candidate serum markers for liver cancer and smear-negative tuberculosis separately.Part one:Verify SAA1 as a serum biomarker of hepatocellular carcinoma by multiple omics strategyObjective:To detect the differently expressed factor in common between HCC cell lines and HCC serum by using transcriptomics and quantitative proteomics. Moreover, analyzing the possibility of SAA1 as the serum marker for liver cancer.Methods:Applying transcriptomics to find out the differentially expressed genes in cancerous liver cell (Smmc-7721) and normal liver cell (HL-7702). Quantitative proteomics was used to select the differentially expressed proteins in serum from 10 healthy subjects and 10 cancerous subjects. The differently expressed factor in common were found by using these two methods. RT-PCR verified the the expressed levels of different factors in cancerous cells and normal cells.107 cases of serum were collected (52 were healthy subjects,55 were subjects with cancerous liver,13 were subjects with HCC metastasis), and they were confirmed by ELISA. Finally, ROC was built and analyzed.Results:Transcriptomic sequencing showed 7 genes were involved in HCC metastasis and inflammation. Quantitative proteomics found 14 proteins were related to HCC metastasis and inflammation. Serum amyloid A1 (SAA1) was the only common protein that was identified by transcriptomic sequencing and quantitative proteomics, which was related to both HCC metastasis and inflammation. RT-PCR showed that the mRNA level of SAAl in cancerous liver cells was 3.14 times higher than normal liver cells (P＜0.0001). ELISA tested that SAA1 in serum from HCC metastasis was 53.24±22.84 μg/mL, non HCC metastasis was 34.08±19.89 μg/mL, and the healthy control was 25.23±13.58 μg/mL. The serum level in HCC metastasis was significantly higher than healthy control (P＜0.01). The area under ROC curve between HCC metastasis and non HCC metastasis was 0.850 (P<0.001). The sensitivity and specificity of detecting HCC metastasis was 84.6% and 81.0%. Besides, the area under ROC curve between liver cancer and normal control was 0.680 (P＜0.01), which had the sensitivity of 63.6% and the specificity of 53.8%.Conclusion:Combining transcriptomic and quantitative proteomics was a quick and efficient method to explore serum protein markers. Our results indicated that SAAl was an inflammatory factor, which may initiate liver cancer and HCC metastasis.Part two:Verify FCN3 as a serum biomarker of smear negative pulmonary tuberculosis by MRM strategyObjective:Verifying the potential protein markers of pulmonary tuberculosis by MRM; and analyzing the possibility of FCN3 as a serum protein biomarker for pulmonary tuberculosis.Methods:The serum often cases of healthy controls SPP-TB,10 cases of pneumonia, and 10 cases of SNP-TB were gathered for MRM analyzing. Relative quantitative analysis was adopted to analyze 5 candidate protein markers of pulmonary tuberculosis (A1AG2, FA9, VTN, SAA4, and FCN3) at the same time. ELISA was applied for FCN3 that had a same result from previous serum proteomics screening. Receiver operating characteristic curve(ROC) was drawn and analyzed for FCN3.Results:The identification of peptides in 5 tuberculosis candidate serum protein markers were detected by MRM, and the separation and peak shape met the requirement of analysis. The results of relative quantitative analysis showed that the 5 serum candidate protein were much lower in patients who had smear negative pulmonary tuberculosis compared to the patients who were healthy (P<0.05). The outcome from ELISA indicated that the concentration of FCN3 was 1060.71±196.10 ng/ml in SPNT,1219.63±246.34 ng/ml in pneumonia, and 1267.24± 260.10 ng/ml in healthy controls. FCN3 was found to be significantly decreased in patients who had smear negative pulmonary tuberculosis than normal controls (P=0.001), and this data matched the result from MRM. Examining FCN3 by ELISA, the area under ROC curve was 0.929 (P＜0.001), and the threshold was 1152.03ng/ml. The sensitivity of diagnosis was 85.61% and the specificity of diagnosis was 82.52% in smear negative pulmonary tuberculosis.Conclusion:MRM spectrum detection provided a high-throughput screening method, which could detect multiple proteins at a same time. It had a great potential on validating the serum marker. Proving by different tests, the decline of FCN3 in SNPT serum suggested FCN3 to be a possible serum candidate biomarker for smear negative pulmonary tuberculosis. |
PDF Full Text Request