Screening And Clinical Study Of Specific Protein Biomarkers For Severe Secondary Pulmonary Tuberculosis

Part one.Analysis of risk factors associated with the death of patients with tuberculosis accompanied by respiratory failure and mechanical ventilation Aims:To observe the clinical characteristics of death tuberculosis patients with respiratory failure requiring mechanical ventilation,and identify factors contributing to death in these patients.Provide early warning and prognostic indicators for progression,aggravated and death to pulmonary tuberculosis complicated with respiratory failure.Methods:A retrospective study of 120 patients admitted consecutively to the medical ICU of Beijing Chest Hospital from January 2010 to May 2014 was conducted.The patients were divided into survival group and death group according to clinical outcome.Collect the datas of demographic,clinical,radiological,bacteriology,and laboratory examinations.A univariate analysis was performed to all factors,then a multivariate analysis was performed to identify risk factors for death.A predictive fatality score was determined.The receiver operating characteristic curve(ROC)was used to analyze the fatality score model.Results:1.The patients' median age was 58.67±20.812 yr(18.0–92.0 yr)and there were 88males(73.33%).The fatality rate was 49.2%.The demographic datas were no significantly difference between the survival group and the death group,(P>0.05).2.The septic shock,renal failure,fungal infection,respiratory rate,C reactive protein,lactate,creatinine,Apache II score were significantly higher in death groupthan in survival group when admitted to ICU.The temperature,mean arterial pressure,BMI and albumin were significantly lower than those in survival group when admitted to ICU,(P<0.05).3.The inspired oxygen concentration,increased mean arterial pressure were significantly higher in death group than in survival group.PH,Pa O2,Pa O2/FIO2 and increased Pa O2 were significantly lower than those in survival group after mechanical ventilation 24 hours,(P<0.05).Other indicators were no statistical different between the two groups,(P>0.05).4.Independent predictive factors of fatality included the Apache II score,initial respiratory rate,lactic acid,C-reactive protein and Pa O2/FIO2 at 24 hours after mechanical ventilation.Fisher discriminant function of the death risk grade of pulmonary TB that required mechanical ventilation with respiratory failure was Z=0.54*Pa O2/FIO2(24 h)-0.32*Apache II score-0.71*respiratory rate-0.39* Lactic acid-0.08*C-reactive protein.The discrimination value was Z=31.56.The AUC was0.903,sensitivity was 72%;specificity was 95.6%.Conclusion:Tuberculosis patients with respiratory failure necessitating mechanical ventilation have a high fatality rate and poor prognosis,particularly those with high Apache II score,high respiratory rate,high lactic acid and C-reactive protein levels,and low Pa O2/FIO2 at 24 hours after mechanical ventilation.The prediction model will help assess the risks of death in patients with tuberculosis and respiratory failure.Part two.Proteomic study of severe secondary pulmonary tuberculosis Aims:To screen distinct proteins in severe secondary pulmonary tuberculosis(STB)by using LC-MS/MS-based label-free quantitative proteomic technology.Analysis the features and functions of proteins and identify different expressed proteins in STB patients.supply new biomarkers for understanding of pathogenesis in TB progression.And Find new protein biomarkers for early warning,prediction of progression and exacerbation of STB.Methods:1.Subjects were prospectively enrolled and recruited by the staff at the Beijing Chest Hospital between March and October 2015.9 STB patients,9 mild secondary pulmonary TB(MTB)patients and 9 non-mycobacterium tuberculosis infected healthy controls(NC)plasma specimens were enrolled in this study.2.3 STB patients,3 MTB patients and 3 cases of NC plasma extraction were collected for identified different expressed proteins using LC-MS/MS-based label-free quantitative proteomic technology.Data was analyzed using gene ontology(GO)and Kyoto Encyclopedia of Genes and Genomes(KEGG).3.Potential biomarkers were further validated by Western blot with 6 STB patients,6 MTB patients and 6 cases of NC plasma.According to difference times,related to TB,inflammation and immunity,and available commercial kits for identified proteins,selected differentially expressed proteins finally.Result:1.399 significantly differentially expressed proteins were screened in plasma of TB,of which 232 proteins were upregulated and 167 proteins were downregulated.153 significantly differentially expressed proteins were screened in plasma of STBgroup,of which 82 proteins were upregulated and 71 proteins were downregulated.2.GO analysis indicated that STB differentially expressed proteins were most in cell part(36.3%)and organelle(27.5%)in the cellular component analysis.Molecular function were most related to protein binding(36.6%)and catalytic activity(26.2%).In the biological process analysis,these proteins were mainly involved in metabolic process(23.2%)and cellular process(18.8%).KEGG analysis showed that the most enriched pathway is focal adhesion pathway.STRING protein analysis revealed that the differentially expressed proteins interact with each other in a physical or functional manner and form a complex network of interactions.3.According to difference times,related to TB,inflammation and immunity,and available commercial kits for identified proteins,we selected 8 differentially expressed proteins validated by Western blot.Upregulation of alpha-1-acid glycoprotein 1 precursor(ORM1),alpha-1-acid glycoprotein 2 precursor(ORM2),interleukin-1 family 6(l L1F6 6),protein S100-A9(S100A9),collagen alpha-1(III)chain preproprotein(Collagen III),probetacellulin precursor(BTC),and downregulation of superoxide dismutase 1(SOD1)and tyrosine-protein kinase isoform B(LYN)in STB,as compared with MTB and NC were noted.The expression of LYN level by Western blot was different from Label free quantitative proteomics method,The expression of BTC level was matched proteomics method,but there was no significant differences between severe secondary pulmonary tuberculosis group and mild pulmonary tuberculosis group.The other 6 proteins expression trend matched with what had been previously observed using the label-free quantitative proteomics method.S100A9,ORM2,IL1F6 and SOD1 were used as the differentially expressed proteins of STB for the subsequent research base on available commercial kits.Conclusion:The 153 differential expression proteins in STB patients was detected by Label free quantitative proteomics technique.They were most in cell part and organelle.Molecular function were most related to protein binding and catalytic activity.In the biological process analysis,these proteins were mainly involved in metabolic process and cellular process.KEGG analysis showed that the most enriched pathway is focal adhesion pathway.The differential proteins had higher catalytic activity and binding characteristics.Through multiple signaling pathways including focal adhesion pathway participated in oxidative stress,inhibition apoptosis and pulmonary fibrosis.It player an an important role in the progression of secondary pulmonary tuberculosis.Part three.Clinical value of plasma differential proteins in predicting the progression and death of secondary pulmonary tuberculosis Aims:To observed the role of four different expressed plasma proteins(S100A9,ORM2,IL1F6,SOD1)in the aggravation of pulmonary tuberculosis.and evaluated its clinical value in predicting the progression and death of STB.Provided the basis for explain the mechanism of tuberculosis progression,and provided new methods and indicators for early warning and prediction of progression and exacerbation of STB.Methods:1.Subjects were prospectively enrolled and recruited by the staff at the Beijing Chest Hospital between March and October 2015.72 STB patients,71 MTB patients and 41 NC plasma specimens were enrolled in this study.2.Detected differentially expressed proteins by enzyme linked immunosorbent assay(ELISA)technique,compared the change trend of differentially proteins between ELISA,Label free and Western blot.Compared related factors by using spearman correlative analysis.ROC curves were used to assess the clinical value of prediction and evaluation of STB.3.The Classification and Regression tree(CART)was used to establish the model.(1)Selected 50 STB,51 MTB patients with 29 NC randomly to establish the CART model,use ROC curves assessed the clinical value of model.(2)Selected 21 STB,21 MTB patients with 12 NC randomly to blind tested the value of model.(3)Randomly selected 85 secondary tuberculosis patients,divided into model STB patients groups and model MTB patients groups by CART model.Followed up the changes of clinical indicators on admission,and at the end of 1,2 and 6 months in35 model STB patients and 50 model MTB patients,then evaluated the early warning effect of this model.4.The 143 patients were divided into survival group and death group according to clinical outcome.Observed the level of four different proteins in two groups.ROC curves were used to assess the clinical value of Predict death.Use CART established the death model,and use ROC curves assessed the clinical value of death model.Result:1.S100A9,ORM2 and IL1F6 expression levels were higher,and SOD1 was lower in MTB patients compared with controls.S100A9,ORM2 and IL1F6 expression levels were higher,and SOD1 was lower in STB compared with MTB patients,(P<0.05).Through correlation analysis,we found S100A9 was positively correlated with erythrocyte sedimentation rate(r=0.268,p=0.001)and negatively correlated with Pa O2(r=-0.398,p<0.001).ORM2 was positively correlated with C reactiveprotein and erythrocyte sedimentation rate(r=0.183,p=0.023;r=0.285,p<0.001).IL1F6 was negatively correlated with albumin(r=-0.182,p=0.031).SOD1 was positively correlated with Pa O2/FIO2 and Pa O2(r=0.307,p=0.043;r=0.268,p=0.032).By using four protein served as a biomarker to discriminate between STB and MTB separately,the AUC were 0.525-0.891,accuracy from high to low were S100A9,ORM2,SOD1 and IL1F6.2.The model for discriminate TB from MTB and NC were generated using CART.(1)The sensitivity,specificity,and accuracy were 91.48% and 90.38%,and 0.941 respectively in discriminating STB from MTB.(2)The sensitivity and specificity were 80.96% and 90.48% respectively in discriminating STB from MTB by blind testing.(3)8 cases of STB and 13 cases of MTB were lost to followed up of 85 patients that distinguished by CART model.27 model STB patients and 37 model MTB patients were included in this study.Erythrocyte sedimentation rate,C reactive protein,WBC and Pa CO2 were significantly higher and lymphocyte,hemoglobin,albumin,Pa O2/FIO2,Pa O2 were lower in model STB compared with model MTB patients.after anti-TB chemical therapy,the sputum negative rate,lesion absorbing rate,cavity closing rate were lower in model STB group at the end of 1,2 and 6 months.Fatality rate,albumin at the end of 1 month,C reactive protein at the end of 1 and 2 months were higher in model STB group than model MTB group,(P<0.05).3.The fatality rate was 49.2% in 143 TB patients.The fatality rate and 1 month fatality rate were significantly higher in STB patients.S100A9,ORM2 and IL1F6 expression levels were higher,and SOD1 was lower in death patients compared with survival group.By using four protein served as a biomarker to discriminate between death and survival groups separately,the AUC were 0.436~0.767.(S100A9,0.767,SOD1 <0.5).4.Built warning death model with four differentially expressed proteins.The sensitivity was 83.21%,specificity was 78.86,and AUC was 0.861.Conclusion:1.The plasma S100A9,ORM2 and IL1F6 levels were higher and SOD1 level was lower in STB compared with MTB and NC groups.Four differentially expressed proteins might related to inflammatory injury and hypoxemia.They had the early warning value for the progression of STB,and S100A9 had the best warning value in four proteins.2.The combination of plasma S100A9,IL1F6,ORM2,and SOD1 levels could achieve higher sensitivity,specificity and accuracy to evaluated STB patients than single protein.It could reflected the clinical features,and had early warning effect on the progression and prognosis of STB patients.3.S100A9,ORM2 and IL1F6 expression levels were higher,and SOD1 was lower in death TB patients.It suggested that the expression levels of these proteins might be related to death.The combination of plasma S100A9,IL1F6,ORM2 and SOD1 levels could had good early warning effect on death than single protein.