Use of Mouse embryonic
stem cell specific microRNA or mRNA signatures to isolate adult
human stem cells from tissues and organs
Specific Aims of the
Project:
Aim 1: We
have identified several microRNA expression clusters during mouse
embryonic stem cell differentiation. We will validate these data
in stem cells by qRT-PCR and northern hybridization.
Aim 2: Promoters
of stem cell specific miRNAs will be isolated through genomics,
bioinformatics and PCR based genome walking techniques.
Aim 3: Stem cell specific miRNA promoters will
be cloned in to reporter genes (GFP) and use as markers to isolate
adult mouse stem cells from complex cell populations. Next, we
will identify cell surface markers from adult mouse stem cells
to screen human counterparts¡¯ from organs and tissues.
Commercial Importance
and rationale of the project:
The rationale of this study
is to utilize early embryogenesis specific miRNA/mRNA expression
signatures (markers) to identify and extract adult stem cells
from human organs. Adult stem cells are important for diseases
such as Parkinson, Huntington, Type 1 diabetes and many more other
incurable diseases. Until today, there has not been an efficient
way to isolate adult stem cells from human organs.
Identification of adult stem cells is important a) To use patient¡¯s
own stem cells to cure diseases and avoid possible immune rejection
effects b) To avoid social and ethical issues that associated
with the use of human embryos for research purposes.
We believe that promoter reporter constructs, early and late embryonic
stem cell specific surface markers, optimized stem cell culturing
protocols that we develop during this study will be able to generate
substantial revenue for further research and development of this
program.
We have identified several
microRNA expression clusters during mouse embryonic stem cell
differentiation. We will validate these data in stem cells by
qRT-PCR and northern hybridization.
Background and Significance:
microRNAs (or miRNAs) are
small non-coding RNAs (21 to 25 nucleotides) that are processed
from large hairpin RNA precursors and are believed to be involved
in a wide range of developmental and cellular processes, by either
repressing translation or triggering mRNA interference (RNA interference).
Over 200 of distinct genes encoding microRNAs have been identified
through either computer-assisted approaches or cDNA cloning strategies
in many organisms including worm, plants, flies, mouse and human
(Lai 2003., Bartel 2004). Recently we have developed a microarray
based robust method to profile microRNA expression in organs cell
lines and tissues in mammalians (Sun and Perera 2004). Using this
method we have identified a group of miRNAs preferentially expressed
in human primary adipocytes and knocking down one such miRNA (miRNA
143) reverse the differentiation process (Esau 2004). We also
reported that a group of kidney specific miRNAs share evolutionary
conserved phylogenetic foot print Ets1 in the upstream of the
miRNA, possibly important for kidney physiological maintenance
(Sun and Perera 2004). Here I discuss a detail protocol for microRNA
global gene expression profiling through end labeling for micro
array experiment.
Embryonic stem cells (ES) are totipotent cell lines derived from
the inner cell mass (ICM) of the mammalian blastocytes (Smith,
2001). In vitro differentiation of ES cells recapitulates some
of the global methylation that takes place shortly after implantation
and has been used to study the epigenetic events that accompany
X chromosome inactivation during midblastula (Wutx and Jaenisch,
2000). The therapeutic potential of stem cells and nuclear cloning
has let to renewed interest in classical models in regeneration.
This long standing problem is undergoing a new beginning spurred
by the availability of new techniques that finally allow analysis
on the cellular and molecular level. miRNA global gene expression
profiling is one such new technique developed to identify stem
cell specific miRNAs (Sun and Perera 2004).
Preliminary Data
Chip design:
We have developed two protocols to profile miRNA global expression
in mouse and human embryonic stem cells. The data presented here
is mainly coming from the custom miRNA array we developed with
the collaboration of Affymetrix Corporation at Mountain View California.
Total of 278 mirs (probes) are being synthesized on Affymetrix
custom chip and modified T4 RNA ligase mediated protocol was developed
to label target miRNA for chip hybridization.
For each PM probe, we have
created ~1000 random mismatch probes with 1 ¨C 5 substitution bases.
We then computed Perfect match (PM) and mismatch (MM) probe qualities
by intercept (measurement of intensity at zero concentration)
and DeltaG (measurement of sequence similarity). The delta-G measurement
is a measurement of sequence similarity. If the delta-G ratio
is 1, then that mismatch would be expected to respond to a PM
spike just as a PM probe and where a delta_G ratio nearer to zero
indicates a probe that won't respond to the PM no matter what
the concentration of the PM spike.
We defined the intercept difference
(InterceptDiff) by Intercept (PM) ¨C Intercept (MM) and DeltaG
difference (DGDiff) by DeltaG (PM) ¨C Delta G (MM). Finally we
group the probes in bin 0 (with -0.25 <InterceptDiff <0.25)
by the number of mismatches and selected the highest DFDiff probes
for 1MM, 2MM, 3MM, 4MM 5MM respectively, i.e. most discriminated
MM probes from PM. 1MM means a probe has one mismatch base, and
2MM is a probe with two mismatch bases, etc. The bin zero MM probes
have about the same zero concentration intensity as the PM probes
and the high DGDiff values to make sure that the MM probes intensity
is much lower than the PM intensity. Please see in the supplemental
material file MM26_all.txt where contains all the MM probes (with
length equal to 26, only one PM) and the file MM26_final.txt lists
the final 25 MM probes selected using methods described above.
In general, we did not tile
probes which cross-hyb to hard prune sequences, as any signal
from those probes is highly likely to result from hybridization
of the repetitive element. In few occasions, we tile probe sets
which cross-hybridize to one of Affymetrix human library sequences
if we could not find a good unique set. For your initial set of
sequences Affymetrix software designed odd-length probes, so there
are files for 17mers through 25mers, although we can tile even-length
probes on the array. For each length, there are two files, a *.hp.all
file and a *.xhy.all file (please see supplemental material).
The first file (hp.all) contains the sequences which cross-hybridize
to any of our hard prune elements, which consist of repetitive
elements such as alu or ERVs, simple repeats, and abundantly expressed
RNAs such as rRNA. The second file (xhy.all) contains the sequences
which cross-hybridize to either one of the other design sequences
or to a sequence in our human library. The format of both of these
types of files is described in section 4.3.2 of Affymetrix Expression
Design Guide (click
here for details).
Target labeling protocol:
10 ¦Ìg of total RNA was first dephosphorylated with shrimp alkaline
phosphatase (SAP), in 37oC for 40 minutes, and then followed by
an inactivation of SAP at 65oC for 40 minutes. This dephosphorylated
RNA product was then end-labeled with biotin donor compound (NEN
biotin-N4 ¨C CTP, NEL-510) by T4 RNA ligase in the presence of
40% Polyethylene glycol (PEG). The reaction was carried out at
37oC for 2 hours. Chip hybridization, processing and scanning
was done according to Affymetrix protocols.
Preliminary data on
stage specific miRNA expression in mouse embryos:
Total RNA from day 0, day 3, day 6, day 9, day 12, and day 15
of the differentiated stem cells was isolated and labeled for
chip hybridization. Results of the raw signal intensity values
were normalized to chip medium 1 and further analyzed by S-Plus
and SAS (statistical packages) to obtain statistically significant
results. Affymetrix custom miRNA chip contains 15 probe replicates
and 25 different mismatches for each miRNA to provide maximum
statistical significant data. Due to the large number of MM probes
to a given PM probe in this chip, one can clearly discriminate
PM from MM and therefore, results we obtain from this method is
highly statistically significant. Here we focus mainly on up and
down regulated miRNAs in differentiated (15 days) compared to
non differentiated (0 days) ES cells. Table 1 illustrates the
expression changes (fold changes) and P-values of several miRNAs.
miRNA 322 shows the highest fold change differences in day 15
compared to day 0. Another interesting observation is the expression
of Let-7a, Let-7e, and Let-7e during the early stem cell differentiation
(Figure 1). Let-7 group of miRNA has been known to be involved
in developmental regulation in C.elegance.
Table 1. Signal intensity values, P-values and expression changes
of miRNA expression during stem cell differentiation in early
embryogenesis.
Figure 1:
Statistically significant miRNA array data was further cluster
(K-mean) to identify stage specific expression in early embryogenesis.
We were able to identify several stage specific miRNA clusters
that are unique for a particular stage but not the others. Figure
2 illustrates a highly expressed miRNA cluster (mir-152, mir-138,
mir-22, mir-357, Let-7a, and Let-7d) in day 15 compared to day
0 during mouse embryogenesis.
Figure 2:
miRNA expression clusters (K-mean)
Aim 1: ¡°We
have identified several microRNA expression clusters during mouse
embryonic stem cell differentiation. We will validate these data
in stem cells by qRT-PCR and northern hybridization¡±
qRT-PCR and Northern validation:
So far the Northern analysis has been the most efficient way of
validating miRNA array data. Take 10 ¦Ìg of total RNA to run on
the polyacrylamide gels and design antisense oligonucleotides
as labeling probes. Follow the protocol according to Ambion (www.ambion.com).
Use U6 labeled probe as the positive control. Some of the data
we validated so far gave us promising results (Figure 3). Expression
of pri-mir also could be an indicator for mir expression and therefore,
qRT-PCR can be used for this purpose. Primer probes are made to
the region where miRNA sequence is located in the pre-mir.
Aim 2: ¡°Promoters of stem cell specific miRNAs
will be isolated through genomics,
bio-informatics and PCR based genome walking techniques¡±
First we will write a PERL
script to detect upstream genomic sequences of selected miRNA.
PERL Script will use the miRNA sequence information and then locate
the corresponding genomic sequence (2-5Kb upstream sequences)
as the putative miRNA promoter from the publicly available human
genome sequences. PCR based genome walker technique will be used
to isolate the upstream genomic fragment. Genome walking protocols
are according to BD Biosciences (www.BD.com)
Aim 3: ¡°Stem
cell specific miRNA promoters will be cloned in to reporter genes
(GFP) and use as markers to isolate adult mouse stem cells from
complex cell populations. Next, we will identify cell surface
markers from adult mouse stem cells to screen human counterparts¡¯
from organs and tissues¡±
For more information please
write to invest@GBNweb.com
