Learning Center
Plans & pricing Sign in
Sign Out



									TSINGHUA UNIVERSITY                                                         Research Experience
Beijing 100084 P.R.CHINA                                                      Rongrong Zhou

                           Research Experience
 Jul., 2004 – present:
Graduate student, Department of Biological Science and Biotechnology (DBSB), Tsinghua
University (THU); Microarray & Bioinformatics Department in National Engineering
Research Center for Beijing Biochip Technology (NERCBBT)
Master’s thesis: Detection mRNA of placental origin in maternal plasma by cDNA microarray
Supervisor: Liang Zhang and Yuxiang Zhou
Backgroud: Noninvasive prenatal diagnosis is a long-sought goal in human genetics. Recent
interest in cell-free DNA in plasma and serum has led to the discovery of fetal DNA in
maternal plasma. This noninvasive source of fetal nucleic acid has already been shown to be
clinically valuable in the prenatal investigation of many conditions. Moreover, a number of
investigators have shown that in addition to DNA, RNA is also present in the plasma of
human subjects, particularly those with cancer. Direct evidences have shown that the placenta
is an important source of fetal nucleic acid release into maternal plasma by demonstrating that
mRNA transcripts from two placenta-expressed genes, namely the genes coding for human
placental lactogen (hPL) and the subunit of human chorionic gonadotropin (hCG), are
readily detectable in maternal plasma. Limited by the throughput of real-time quantitative
RT-PCR which is currently used in this field, it’s difficult to detect many mRNA transcripts
of placenta-expressed genes simultaneously.
However, the advent of microarray technology will allow the detection of a great deal of these
RNA and the very rapid development of new plasma RNA markers, which can be used for
prenatal monitoring.
Methods: In this project, we separate the plasma from maternal blood by centrifuge and pass
it through filter with pore sizes of 5m. Then LS TRIzol was added to extract RNA. In the
following process, RNA was amplificated by two-step cRNA synthesis and labeled with Cy5
or Cy3 by Klenow enzyme after reverse transcription. Finally, the labeled targets were
hybridized with our cDNA chips and the arrays were scanned with a ScanArray Express

 Sep., 2003 – Jun., 2004:
Graduate student, DBSB, THU; Microarray & Bioinformatics Department in National
Engineering Research Center for Beijing Biochip Technology (NERCBBT)
Master’s thesis: Genome-wide oligonucleotide microarray analysis of gene-expression
profiles in placentas of preeclamptic pregnancies

TSINGHUA UNIVERSITY                                                         Research Experience
Beijing 100084 P.R.CHINA                                                      Rongrong Zhou

Supervisor: Liang Zhang and Yuxiang Zhou
Backgroud: Preeclampsia is a multisystem disorder specific to human pregnancy which is
characterized by gestational hypertension, fluid retention, and proteinuria. The hypertensive
disorders of pregnancy affect up to 8% of all gestations. Although preeclampsia has been
recognized as a pregnancy disorder since antiquity, the pathogenesis is not understood despite
decades of research and the syndrome of preeclampsia is still one of the major pregnancy
disorders and the leading cause of pregnancy-related maternal and fetal morbidity and
Microarray analysis is a technology that allows the characterization of a whole-genome
expression by showing the relative transcript levels of thousand of genes in one experiment,
thereby speeding the identification of differential expression profiles. At present, large-scale
gene expression studies based on microarray analysis have been applied to observe the gene
expression patterns in human placentas from pregnancies complicated by preeclampsia in the
third trimester. However, a whole genome-scale gene expression study has not been applied
up to now to observe the gene expression patterns in human placentas from pregnancies
complicated by preeclampsia. Clearly, further insight is required in this area given that the
overall result of the candidate gene studies are still rarely in accordance with each other.
Methods: Placentas tissues from five normal pregnancies and five pregnancies complicated
by preeclampsia were collected. mRNA level of five preeclamptic placentas were examined
respectively using a genome-wide 70-mer oligonucleotide microarray (21,329 human genes)
in comparison with the pooled control consisting of total RNA from five normotensive
placentas after the control cDNA and experimental cDNA were labeled by Cy5 and Cy3

Results: Over one hundred genes were found consistently down or up-regulated in at least
four experiments. Many genes related to unbalance of reactive oxygen metabolites in placenta,
abnormal trophoblast invasion, disorders of lipoprotein metabolism and signal transduction
were among the differentially expressed genes, some of which had been reported to have close
correlation to the pathology of preeclampsia. Most results of microarray analysis were also
confirmed by real-time PCR.

 Jan., 2004-Apr., 2004
Graduate student, DBSB, THU; NERCBBT
Research of biostatistics and bioinformatics
Supervisor: Liang Zhang and Yuxiang Zhou
Still with its high-throughput, microarray brings out thousands and ten thousands data. How to
manage and analyze so many large and complex data effectively is a great challenge to
bioinformatics. It is inevitable to have random error from various origins in the process of
microarray. How to distinguish and remove these errors is the prerequisite to get the correct

TSINGHUA UNIVERSITY                                                             Research Experience
Beijing 100084 P.R.CHINA                                                          Rongrong Zhou

result. We analyzed the errors from available microarray data and then put forward a new kind
of error model and a robust algorithm to estimate the parameters of the model. Applying to
real and simulated data, this model and algorithm were proved to be correct and efficient.
Lastly, combining this error model with Z-testing, a full statistical model was set up for data
analysis, to select the differentially expressed genes from the result of microarray. Except for
random error, there is also systematic error in the experiment of microarray. We can get the
correct fold change of gene expression by normalization to eliminate this systematic error,
between samples. We compared commonly used normalization methods and then developed
an improved and robust algorithm, which we called malm, based on linear regression analysis.
It was proved that malm could be effectively applied to real and stimulated data. Especially,
based on its stability, malm can also be used to normalize biased expression data of

 Sep., 2002 – Jul., 2003:
Graduate student, DBSB, THU; HLA typing Department in NERCBBT
Supervisor: Huafang Gao and Yuxiang Zhou
During this period, I participated in the project of HLA typing chip, the principle of our
experiments can be described by the flow chart below. We finally designed proprietary probes
covering A, B, and DRB1 loci and achieved superior accuracy that exceeded the national

    Target Gene            Purification & Quantification       Printing DNA
    Amplification                of PCR Products                Microarrays


       Sample Genotypes &                     Automated Data              Image Scanning & Data

         Serological Types                        Reporting                    Processing

 Feb., 2002 – Jun., 2002:
Undergraduate student, Department of Biophysical Science and Technology, Nankai
Bachelor’s thesis: The inhibition of rhein on increase in [Ca2+]i and TNF-release in
lipopolysaccharide (LPS)- stimulated macrophages
Supervisor: Yang Wenxiu
Methods: TNF- production was determined using TNF-sensitive L929 cell line and by
MTT method and the kinetics of intracellular free Ca2+ in a single cell were monitored
continuously by confocal laser scanning microscope.

TSINGHUA UNIVERSITY                                                        Research Experience
Beijing 100084 P.R.CHINA                                                     Rongrong Zhou

Results: Rhein had not a distinct effect on TNF-production in normal PM,but could
obviously inhibit TNF-release in LPS-stimulated PMin dose-dependent manner, 10-4
mol/L Rhein could inhibit the effect of LPS by 72%. Noticeably, Rhein not only lowered the
rising of [Ca2+]i in LPS-stimulated PM, but also altered the dynamical mode of [Ca2+]i, which
means that Rhein could transform the ‘plateau-shaped’ peak of [Ca2+]i into the oscillating
mode. In the case of extracellular free Ca2+, Rhein further transformed the [Ca2+]i dynamics
into a single peak of lower amplitude. These results indicate that Rhein reducing the increase
in [Ca2+]i is the key of signal transduction pathway that Rhein inhibited the TNF-production
in LPS-stimulated PM It further suggests that inhibition and quasi-periodic modulation on
influx of extracelluar calcium as well as on intracellular calcium release is involve in the
regulative mechanism of Rhein.

National Nature Science Foundation of China
Noninvasive Oligonucleotide Microarray for Predicting PIH
Qianyong Zhu, Zhi Li, Liang Zhang, Changxu Xu, Bing Chen, Rongrong Zhou and Yong
The Third Military Medical University, China & NERCBBT

No: 30400480(2004)

Abstract: Pregnancy-induced hypertension syndrome (PIH) is a common pregnancy
complication, which is a main cause of maternal and newborn morbidity and mortality. Prior
to the emergence of clinical symptoms, early prediction and diagnosis and subsequent
intervention in time can decrease the incidence and slow the process of this disease greatly.
However, all of the predictors currently used in research cannot meet clinical requirements.
This project adopted prospective cohort study design. Followed comparison of gene
expression pattern in plasma of normal pregnant women and those of PIH patients finally
diagnosed in early, midterm and late period of pregnancy was conducted using human total
genomic oligonucleotide microarray representing over 23000 pieces of genetic information.
Combined bioinformatical methods with comprehensive analysis of clinical data, multiple
corresponding genes that can predict the onset of PIH were screened, which were
retrospectively confirmed using Real-Time PCR. A fast, accurate, and noninvasive
oligonucleotide microarray for predicting PIH was preliminarily devised for screening
clinically high-risk or nomal pregnant women. The study is helpful for further investigation in
preventing and radical cure of PIH.


To top