| In recent years, many studies demonstrated that lipids, especially sphingolipids are highly bioactive compounds that serve not only as core components of biological structures, such as membranes and lipoproteins, but also as regulators of cell proliferation, differentiation, cellular interactions and migration, intracellular (and extracellular) signaling, membrane trafficking, autophagy, and cell death. Therefore, lipids have become the new focus of biochemistry. Meanwhile, development of new analytical platform for lipids has become the hotspot in the field of analytical chemistry. With the development of modern mass spectrometry, especially mass spectrometry coupled to liquid chromatography, more structural details of lipids have been revealed, which make the connection between structure and function clearer. Based on the strategy of targeted metabolomics, lipidomics was introduced as the large-scale study of pathways and networks of cellular lipids in biological systems.Natural abundance of lipids varies a lot in biological system, ranging from nanogram scale to microgram scale. Among those lipids, sphingolipids belong to one subclass of lipids whose abundance is among nanogram scale. With relatively low ionization efficiency, sphingolipids should be detected in a more sentive mass spectrometer, eg. triple quadruple mass spectrometer. According to the variance of natural abundance and structural characteristics of those lipids, we have built an integrated lipidomics platform to fulfil the detection and quantification. In this thesis, we first developed a targeted sphingolipidomic platform, consisting of a high performance liquid chromatography coupled with tandem triple quadruple mass spectrometry. This platform was capable of quantification of43sphingolipids, covering the core metabolic network of sphingolipids. In the beginning, we optimized the extraction method by introducing methyl-tert-butyl-ether:methanol:water (20:6:5, v/v/v) solvent system. The sample volume was0.1mL for liquid samples (eg. plasma, serum, tissue homogenates) and1mg protein for solid samples (eg. cell). The absolute recovery was among60-80%. A reversed phase C8column in gradient mode was employed in the chromatography part. Mobile Phase A consisted of2mM ammonium formate in water containing0.2%formic acid. Mobile Phase B consisted of1mM ammonium formate in methanol containing0.2%formic acid. In the mass spectrometry, all target compounds were detected in ESI-positive mode. The approach of segmental multiple reaction monitoring improved the sensitivity and accuracy. At least one internal standard was introduced in each segment. The result of method validation demonstrated that the limit of quantification was1.0pmol/mg protein; linear range:12.5-2000.0pmol/mg protein (for sphingomyelins) and2.5-400.0pmol/mg protein (for other sphingolipids) with linear correlation coefficient greater than0.99; The intra-day and inter-day precision was less than15%; the accuracy was among80～120%; The working solution was stable in room temperature for6hours and in-20°for60days. Above results proved this method was accurate and reliable for investigating the sphingolipid levels in biological samples. In order to fully cover whole metabolomic network of sphingolipids, we further expanded the number of target sphingolipids from43to74and integrated the linear ion trap-Fourier transform ion cyclotron resonance (LTQ-FT) mass spectrometer as a complementation to profile the four categories of highly abundant lipids, including glycerophosphocholines, glycerophosphoethanolamines, diacylglycerol and triglycerides, totally216lipids. The LTQ-FT was working in ESI positive full scan mode. A reversed phase C8column in gradient mode was employed in the chromatography part before the mass spectrometry. Mobile phase A was comprised of0.1%formic acid in water containing2mM ammonium acetate. Mobile phase B was comprised of2-propanol/acetonitrile (2:5, v/v) containing2mM ammonium acetate and0.1%formic acid. The Lipid Data Analyzer(?) was used to analyze the raw data file in a high throughout manner. Partial validation demonstrated that the limit of quantification was0.02nmol/mg protein; the linear range was among0.02-200nmol/mL with linear correlation coefficient greater than0.95; the precision was less than15%; the accuracy was among80%-120%. Above results confirmed that this lipidomics platform was accurate and reliable for investigating the lipidome qualitatively and quantitatively in biological sample.During the method development, we timely applied the method in real sample analysis. First of all, we used the sphingolipidomic method to profile sphingolipids in the plasma, kidneys, livers and spleens of BALB/c mice from four experimental groups:control, delayed-type hypersensitivity (DTH) model, DTH+triptolide, and control+triptolide. Orthogonal partial least squares discriminant analysis (OPLS-DA) was used to identify potential biomarkers associated with variance between groups. Relationships between the identified biomarkers and disease markers were evaluated by Spearman correlation. The sphingolipidomic result revealed marked alterations in sphingolipid levels between groups that were associated with the effects of the disease and triptolide treatment. Based on this data,23potential biomarkers were identified by OPLS-DA, and seven of them correlated markedly with the disease markers (p<0.05). Restoration of proper sphingolipid levels may attribute to the therapeutic effect of triptolide treatment. Furthermore, these findings demonstrate that targeted sphingolipidomic analysis followed by multivariate analysis presents a novel strategy for the identification of biomarkers in biological samples.It is reported that hepatitis viruses rely on constituents of the host cell to provide the energy, macromolecules, and structural organization necessary for survival, and they must cross membranes either by transient local disruption of membrane integrity or by cell lysis. Above processes are related with sphingolipid metabolism. Therefore, we conducted a research on the relationship between serum lipids and chronic HBV infection. The sphingolipidomic platform was used to examine the sphingolipids in156serum samples prospectively collected from two independent cohorts. The training and validation cohorts comprised20and28healthy controls (CTRL),29and23patients with chronic hepatitis B (CHB), and30and26patients with hepatitis B virus related acute on chronic liver failure (HBV-ACLF), respectively. Nine potential biomarkers were found and confirmed between CHB and HBV-ACLF, whereas none was confirmed between CTRL and CHB by multivariate analysis. Such result provided new clues for pathogenesis research for chronic HBV infection. Further, a3-month mortality evaluation of HBV-ACLF patients showed that dhCer(d18:0/24:0) was significantly higher in survivors than in non-survivors (including deceased patients and those undergoing liver transplantation, p<0.05), and showed a prognostic performance similar to that of the MELD score. This preliminary study firstly revealed that the serum sphingolipid composition varies between CTRL and patients with chronic HBV infection and provided new potential method in the prognosis of HBV-ACLF patients.Based on our previous study, we further applied the integrated lipidomics platform on plasma samples collected from patients with chronic hepatitis C virus (HCV) infection. The plasma samples were collected by Beijing YouAn Hospital, including11healthy controls and113patients with chronic HCV infection. All the patients were grouped by their intrahepatic inflammation grade (IG) of liver tissue into IG01, IG2and IG34. As a result, the plasma lipid profile varies a lot between healthy controls and patients with chronic HCV infection (47of117quantified lipids significantly changed). Only eight lipids (HexCer (dl8:1/22:0), HexCer (d18:1/24:1), HexCer (d18:1/24:0), PC(34:4) and PC(40:5)) showed significant differences between mild and severe intrahepatic inflammation grades, indicating that they could be useful as novel non-invasive indicators of intrahepatic IG. Based on these results, we speculate that variations in lipid composition arise as a result of HCV infection, and are caused by HCV-related digestive system disorders rather than progression of the disease.Later on, we further developed an online HILIC×RP2D-lipidomic platform consisted of a3-step strategic approach for screening candidate compounds in an untargeted way and quantitation of those compounds in MRM mode, a targeted way. A triple quadruple mass spectrometry was used in this study. This platform was demonstrate to be especially useful in discrimination of PCs and SMs by resolving those two types of compounds mutually and internally. Mild alkaline hydrolysis, an unstable factor in previous sample preparation was no more needed. The accuracy, precision and limit of detection were evaluated. We further applied this platform in analyzing lipidome of male and female mice plasma obtained from three different positions. Totally,154lipids were correctly quantified in all samples, including1ceramide,1ceramide-1P,1sphingosine,60PCs,19SMs and72TGs. The results demonstrated that male and female mice plasma shows significant distinct lipid profile. The plasma lipidome from retro-orbital sinus/plexus, abdominal aorta and tail vein are also different. Above result provided us a hint that when to repeat a lipidomics experiment, repeating the method of sample collection is very important, which includes not only animal gender but also its originate.This thesis work has built a series of qualitative and quantitative lipidomics platforms. By analyzing actual biological samples, the coverage and depth of them was amply demonstrated. Our results showed that lipids are tightly connected with disease progression, providing novel clues for diagnosis, prognosis and mechanism research. |
PDF Full Text Request