| IntroductionNeural tube defects (NTDs) are quite common congenital central nervous system abnormalities. Embryologically, failure in neural tube closure determines a group of diseases called neural tube defects, including anencephaly, spina bifida, meningocele, myelomeningocele, encephalocele, hydrencephaly, hydrosyringomyelia. Neural tube defects have been described in all popμlations in which they have been sought. Half of these defects are spina bifida and half are anencephaly with or without spina bifida. Historically, the birth prevalence of NTDs has ranged from about 0.5 to 6 in 1,000 births. Both environmental and genetic causes may be the reason of neural tube defects. Prevalence of neural tube defects is about 10.63 in 10,000 births, female is superior to male, countryside is superior to city. Abortion will occur in 54%of neural tube defects before twenty seven pregnancy weeks, and 46%of neural tube defects will entry in perinatal period. Newborns with neural tube defects will be a serious burden to their families and decrease the population quality.At present, prenatal diagnosis may be the most important screening method. Maternal serum alpha-fetoprotein (MSAFP) and amniotic fluid alpha fetoprotein (AFAFP) screening and ultrasonography have been used for detecting neural tube defects prenatally. The improved resolution of ultrasonography has allowed detailed imaging of nervous system anatomy and enhanced understanding of both development and pathology. Prenatal screening in the general population is based on two complementary methods, maternal serum AFP screening and ultrasound screening. It is necessary for ultrasonography that gestation age must excess to 12 weeks in pregnancy, and many factors will result in the missed diagnosis of ultrasonography. Prenatal diagnosis in pregnancies identified as being at a high risk of open neural tube defects requires fetal scanning by an expert sonographer. MSAFP and AFAFP levels will be abnormally high in the open neural tube defects and can be used as screening tests. However, both specificities are not good and such screenings are inadvisable. Amniocentesis for acetylcholinesterase electrophoresis is an invasive method and should be limited.The surface enhanced laser desorption/ionization flight-of-time mass spectrometry (SELDI-TOF-MS) technology, developed from the technology of chromatography and mass spectrometry (MS), includes a protein capture chip. The specific protein can be captured on the chip surface. SELDI-TOF-MS offers the advantages of rapid and simple examination as well as high specificity and sensitivity. It analyses small volumes of clinical samples without destroying the proteins to be detected and is capable of examining proteins and peptides, which are not available for conventional methods. New proteins specific to some diseases and characterization of these proteins can be discovered and captured by comparative analysis of the mass spectra of the samples from patients and normal controls. Thus, it is suitable for examination of small volumes of samples such as serum, urine, amniotic fluid with complicated components, and an ideal method to find protein biomarkers. At present, there are many exciting SELDI-TOF-MS applications, which have been described by numerous laboratories, especially in studying cancers of various organs, including prostate, ovary, breast, lung, colon, and others. To our knowledge, there has been no study reported using SELDI-TOF-MS technology to investigate neural tube defects. In this study, the serum, urine, amniotic fluid proteins of the patients and rat with neural tube defects and those of normal controls were analysed. SELDI-TOF-MS (Ciphergen Biosystems, USA) was used to detect specific protein biomarkers in serum, urine, amniotic fluid and to obtain diagnostic fingerprints which, coupled with the dicision tree analysis patterns for detection of serum and urine proteins, would give an early diagnosis of neural tube defects.Material and methods1. Collection and preparation of samples(1) serum samplesOf the 31 serum samples obtained,17 were from patients with neural tube defects and 14 from healthy volunteers undergoing pregnancy routineμltrasonography examination in the Department of ultrasound, Shengjing Hospital, China Medical University. The diagnosis of the neural tube defects was confirmed routine ultrasonography examination. All fasting blood were obtained early in the morning. Untreated whole blood were collected and allowed to clot 1-2hrs at room temperature. Serum was purified from blood by centrifugation for 10 minutes (4000rpm) at 4℃, aliquoted as 100 microliters and stored at-80℃. Serum samples were not subjected to more than two freeze-thaw cycles before the assay.(2) urine samplesOf the 35 urine samples obtained,20 were from patients with neural tube defects and 15 from healthy volunteers undergoing routine ultrasonography examination in the Department of ultrasound, Shengjing Hospital, China Medical University. The diagnosis of the neural tube defects was confirmed routine ultrasonography examination. All fasting urine were obtained early in the morning. Urine were collected and centrifugated for 5 minutes (2500 rpm) at 4℃immediately after collection, aliquoted as 100 microliters, and stored at-80℃until analysis. Urine samples were not subjected to more than two freeze-thaw cycles before the assay.(3) amniotic fluid samples of fetal neural tube defectsOf the 20 amniotic fluid samples obtained,11 were from patients with neural tube defects and 9 from healthy volunteers undergoing routine ultrasonography examination in the Department of ultrasound, Shengjing Hospital, China Medical University. The diagnosis of the neural tube defects was confirmed routine ultrasonography examination. AF was retrieved by transabdominal amniocentesis. Centrifugation were performed immediately after amniocentesis as part of the clinical workup. Specimen was centrifuged at 2500rpm and 4℃for 5 min, aliquoted as 100 microliters, and stored at -80℃until analysis.The Ethics Boards of China Medical University approved this research study. Written informed consent was obtained from all participants prior to the procedure. Subjects were enrolled prospectively based on the availability of one of the investigators for consent procedures.(4) amniotic fluid samples of fetal rat neural tube defectsTimed-mated (GD 0 assigned as the day of mating) female Wistar rats were acquired from medicine animal center of China Medical University. Rats were healthy and mature and ranged between 220-250g at the initiation of dosing. The rats were housed individually in cages. Certified rodent diet and drinking water were provided ad lib. Animals were randomly assigned to experiment groups and control groups. Environmental conditions were set to maintain room temperature and humidity for rats. Room lighting was on a 12-hr light/dark cycle. The facility is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International. Pregnant Wistar rats were administered with a suspension of all-trans retinoic acid (120 mg/kg) and liquid paraffin (40 mg/ml) on gestational days 10 and control group were studied. Amniotic fluid of fetal rat on gestational days 17 were aspirated by a needle. The deformities of fetal rats were identified by microscopy. Tha rats with spina bifida and anal atresia were chosen as experiment objects. The amniotic fluid must not be muddy, and amniotic fluid was immediately centrifuged at 4000rpm and 4℃for 10 minitus, aliquoted as 100 microliters, and stored at-80℃until analysis.2. Samples preparationThe serum samples (urine, amniotic fluid) from the patients and rats with neural tube defects were collected prior to the start of treatment. All samples were thawed at room temperature and then centrifugated for 2 minutes (10,000 rpm) at 4℃. Ten microlitres of supernatant were first put into 20μl of protein denaturant (9 mol/L urea, 2%CHAPS,1%DTT,50 mmoL/L Tris-HCL, pH 9.0) and shaken at 4℃for 30 minutes on a shaker for complete protein denaturing. Then,30 microlitres of denatured sample was further diluted with 50 mmol/L sodium ethanoate (acetate, pH 4.0)to 390μl (urine to 280μl, amniotic fluid to 200μl)3. Reagents and instrumentsSodium ethanoate (acetate), carbamide, acetonitrile, trifluoroethanoic acid and SPA(sinapinic acid)were purchased from Sigma(USA). Protein chip biosystem(PBSⅡC) and CM 10 chip were purchased from Ciphergen Biosystems (USA)4. samples preparationA protein chip was installed into a bioprocessor (Ciphergen, USA). Two hundred microlitres of binding buffer (50 mmol/L sodium ethanoate, pH 4.0) were added into each spot of an eight spot chip, and shaken for 5 minutes on a shaker. The shaking was repeated in the same way after the binding buffer was removed. One hundred microlitres of sample were added and shaken for 60 minutes. After the serum sample (urine sample, amniotic fluid sample) was thrown off, the chip was washed two times with binding buffer for 5 minutes each time, rinsed once with ultrapure water and airdried. The proteins bound CM 10 chip was then treated with 1μl of saturated sinapinic acid solution(Ciphergen,USA). The chip was airdried for future examination.5. collection and preparation fo dataA mass spectrometer was calibrated with chips that had been bound with all-in-one standard proteins to set up parameters. The parameters used were:the optimal detection mass/charge size (m/z) range was between 1000 and 30,000Da; the laser intensity was set at 225 and the detector sensitivity was set at 9. An average value of 130 spots was presented for each sample. All samples were detected with the same parameters. All the raw data were normalized with the ProteinChip Software version 3.1.1. The peaks m/z of the samples with more than 1000 Da were normalized with biomark wizard of ProteinChip Software version 3.1.1 for noise filtering. The first threshold for noise filtering was set at 5. Generation of the tree analysis pattern was performed by Biomarker Patterns Software (Ciphergen Biosystems, Inc.). The data of spectra were analyzed by bioinformatics tools-Biomarker Wizard and Biomarker Pattern Software.6. statistical analysis of the dataThe data were processed with the ProteinChip Software version 3.1.1 for Mann-Whitney U test. P value< 0.05 was considered statistically significant.Results1.maternal serum protein biomarkersA total of 55 qualified mass peaks (signal-to-noise ratio> 5) were detected in the training set. Compared with the spectra of control groups, there were 12 potential markers detected in the spectra of the neural tube defects patients, the protein expression was high in 8 of which (4105,4297,4188,6650,8583,3282,2750, 3327) and low in the 4 of which(5497,28078,9155,9434). The softwares Biomarker Wizard and Biomarker Pattern Software automatically, under given conditions, selected 2 biomarker proteins (4105,7788) to be used to establish a three layer decision tree differentiate to diagnose neural tube defects and differentiate neural tube defects from control groups with a specificity of 100%and a sensitivity of 88.2%.2.maternal urine protein biomarkersA total of 39 qualified mass peaks (signal-to-noise ratio> 5) were detected in the training set between 1000 and 30,000Da. Compared with the spectra of control groups, there were 5 potential markers detected in the spectra of the neural tube defects patients, the protein expression was high in 4 of which (8320,8209,9099,10567) and low in the 1 of which (3458). The softwares Biomarker wizard and Biomarker Pattern Software automatically, under given conditions, selected 2 biomarker (9096,8244) proteins to be used to establish a three layer decision tree with three terminal nodes differentiate to diagnose neural tube defects and differentiate neural tube defects from control groups with a specificity of 93.3% and a sensitivity of 80.0%.3.maternal amniotic fluid protein biomarkersA total of 35 qualified mass peaks (signal-to-noise ratio> 5) were detected in the training set between 1000 and 30,000Da. Compared with the spectra of control groups, there were 7 potential markers detected in the spectra of the neural tube defects patients, the protein expression was high in 6 of which (14700,7995,15891,16027,13776, 11040) and low in the 1 of which (23417)4.rat amniotic fluid protein biomarkersA total of 55 qualified mass peaks (signal-to-noise ratio> 5) were detected in the training set between 1000 and 30,000Da. Compared with the spectra of control groups, there were 9 potential markers detected in the spectra of the neural tube defects patients, the protein expression was high in 5 of which (11658,27387,7898,11603,13829) and low in the 4 of which (5124,14702,5403,13626)5.comparation between maternal amniotic fluid protein biomarkers and rat amniotic fluid protein biomarkersCompared between maternal amniotic fluid protein biomarkers and rat amniotic fluid protein biomarkers, five mass peaks, including 4138 (4145),7947 (7941), 8588 (8588),9385 (9390),14702 (14700),especially m/z 14700 (P=0.006) may be the same protein/peptide of them.Conclusion1. These discrepancy mass peaks of serum, urine and amniotic fluid identified by SELD-TOF-MS (surface enhanced laser desorption/ionization time-of-flight mass spectrometry) may be the specific serum, urine and amniotic fluid protein biomarkers of neural tube defects.2. New serum and urine biomarkers of neural tube defects have been identified, and this SELDI mass spectrometry coupled with decision tree classification algorithm will provide a highly accurate and innovative approach for the early diagnosis of neural tube defects.
|
PDF Full Text Request