Absolute quantification of gene expression in biomaterials research using real-time PCR

Biomaterials (Impact Factor: 8.56). 01/2007; 28:203-210.
ABSTRACT
Cited By (since 1996): 18, Export Date: 27 September 2011, Source: Scopus

Full-text

Available from: Fook Tim Chew, Jul 17, 2015
Biomaterials 28 (2007) 203210
Technical note
Absolute quantification of gene expression in biomaterials research
using real-time PCR
David Tai Leong
a
, Anurag Gupta
b
, Hui Fen Bai
c
, Guoqiang Wan
d
, Li Foong Yoong
d
,
Heng-Phon Too
d
, Fook Tim Chew
a
, Dietmar Werner Hutmacher
c,e,
a
Department of Biological Sciences, National University of Singapore, Republic of Singapore
b
School of Medical Science and Technology, Indian Institute of Technology, India
c
Division of Bioengineering, National University of Singapore, Republic of Singapore
d
Department of Biochemistry, National University of Singapore, Republic of Singapore
e
Department of Orthopedic Surgery, National University of Singapore, Republic of Singapore
Received 20 June 2006; accepted 5 September 2006
Abstract
One major measurement of tissue-engineered constructs efficacy and performance is determining expression levels of genes of interest
at the molecular level. This measurement is commonly carried out with reverse transcription-polymerase chain reaction (RT-PCR).
In this study, we described a novel method in achieving absolute quantification of gene expression using real-time PCR (aqPCR). This
novel method did not require molecular cloning steps to prepare the standards for quantification comparison. Standards were linear
double-stranded DNA molecules instead of the typical gene-in-plasmid format. aqPCR could also be used to give relative quantification
comparisons between samples simply by dividing the copy numbers readings of the gene of interest with that of the normalization gene.
RNA was extracted from monolayer and from polycaprolactone scaffold cultures and assayed for b-actin and osteocalcin genes. We
compared our aqPCR method with end-point PCR since end-point PCR is still a common means of measuring gene expression in the
biomaterials field. This study showed that aqPCR was a better method to quantify gene expression than end-point PCR. With our
described linear DNA standards method, we were able to obtain not only relative quantification of osteocalcin and b-actin expression
level but also actual copy numbers of osteocalcin and b-actin for the monolayer culture and to be 1.34 10
4
and 1.45 10
7
copies,
respectively and for the scaffold cultures to be 772 and 2.83 10
5
copies, respectively per starting total RNA mass of 10 ng. The
standards curves made from these linear DNA standards showed good linearity (R
2
¼ 0:9964 and 0.9902 for osteocalcin and b-actin
standards graphs), ranged from 10 to 10
9
copies and of comparable accuracy to current absolute quantification real-time PCR methods
(which used plasmid standards obtained through molecular cloning methods). Our method might be a viable and more user-friendly
alternative to current absolute quantification real-time PCR protocols.
r 2006 Elsevier Ltd. All rights reserved.
Keywords: Real-time PCR; Absolute quantification of gene expression; Osteocalcin; Adipose-derived cells
1. Introduction
When analyzing tissue-engineered constructs or studying
biomaterials–cells inter actions, it is almost a certa inty to
quantify perfor mance of such constructs in terms of in
vitro and in vivo characteristics. Measurements are usually
made in the macroscopic and microscopic level. However,
in order to complete the characterization, it is often
necessary to measure performance of these constructs at
the molecular level. It comes as no surprise that recent
papers in the biomaterials and biomedical fields used gene
expressions of cells and their responses to their 3D
environments [1–5] to measure the performance of these
tissue-engineering constructs.
Reverse transcription-polymera se chain reaction (RT-
PCR) is an in vitro method for enzymatically amplifying
(indirectly) defined sequences of messenger RNA (mRNA)
ARTICLE IN PRESS
www.elsevier.com/locate/biomaterials
0142-9612/$ - see front matter r 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biomaterials.2006.09.011
Corresponding author. Tel.: +65 6516 1036.
E-mail address: biedwh@nus.edu.sg (D.W. Hutmacher).
Page 1
[6]. This method is sensitive enough to compare mRN A
from as little as one cell [7]. RT-PCR can be used to
compare levels of mRNA in different sample populations,
to characterize patterns of mRNA expression and to
discriminate between closely related mRNAs [8]. With this
immense ability to amplify rare copies of cDNA, PCR has
opened up another level of sophistication in ascertaining
gene expression in cells–environment interactions.
RT-PCR can be carried out in two ways, end-point PCR
and real-time PCR. The major difference between these
two methods is that for end-point PCR , quantification
takes place at the end of the entire PCR react ions while
real-time PCR takes measurements at the exponential
phase of the PCR.
There are two common categories of quantification using
real-time RT-PCR, (a) relative and (b) absolute quanti-
fication.
Relative quantification is commonly carried by normal-
ization of the expression levels of the gene of interest with
that of the housek eeping genes. Relative quantification
with real-time PCR has been used in comparing osteogenic
genes in bone marrow-cancellous bone [9]; chondrogenic
gene expression of bovine chondrocytes seeded on hybrid
PLGA mesh with collagen-filled pores [10]. Relative
quantification however would only give a ratio of the gene
of interest comparison and not actual copy numbers in a
defined concentration of mRNA population. Useful
biological information might be lost with using the ratio.
Since a high ratio may not necessarily mean a high
expression of the gene of interest as the ratio is sensitive
to the expression level of the normalizing gene (the
denominator).
Absolute quantification relies on a standards plot
constructed from the known concentrations of standards
template and corresponding levels of real-time PCR data.
Commonly, standards are derived from purified plasmid
dsDNA, in vitro-transcribed RNA or in vitro-synthesized
ssDNA [11]. The amount is quantified spectroscopically at
260 nm or with DNA fluorescent dye [12] and converted to
number of copies using the molecular weight of the RNA
or DNA sequence. Actual copy numbers of the gene of
interest in the samples could then be read off the standards
plot. Absolute quantification therefore would eliminate the
ambiguous use of ratios.
However, the current methods of producing standards
for the absolute quantification real-time PCR can be
technically tedious since it often involves molecular cloning
of the target sequ ence into vector systems and amplifica-
tion in Escherichia coli. In this study, we introduced
another method to quantify actual copies of starting
mRNA using real-time PCR technique but did not require
molecular cloning steps to obtain the standards. This
method used linear double-stranded (ds) DNA standards
purified from PCR reactions containing the target
sequence. mRNA from positive control cells was first
reversed transcribed to cDNA and then with appropriately
designed primers, the target sequence was amplified with
standard PCR methods. The correct PCR products were
confirmed using standard gel electrophoresis and isolated
from the gel band. The PCR products could be further
sequenced for further confirmation of the sequence. The
purified ds DNA were then quantified and serially diluted
to produce a standards graph. For each assaying routine,
the samples were also subjected to the same conditions as
the standards. The standards readings were used to plot a
standard plot and the actual copy numbers of the samples
could then be read off the plot.
This technique could be easily adapted to existing real-
time PCR protocols. We also discussed on the advantages
and limitations of the described aqPCR method.
2. Materials and methods
2.1. Monolayer cell culture
Human fetal osteoblasts (hFOBs—CRL-11372 from ATCC), were
maintained in normal culture medium consisting of Dulbecco’s modified
Eagle’s medium (DMEM, Sigma D1152) supplemented with 10% fetal
bovine serum (FBS, Hyclone) at 37 1C, 5% CO
2
and 95% humidity.
Primary adipose tissue-derived progenitor cells were cultured in the above
conditions. Osteogenic culture medium consisted of normal culture media
supplemented with 50 m
ML-ascorbic acid–2–phosphate (Sigma Aldrich,
A8960), 10 m
M b-glycerophosphate (Sigma Aldrich 6376), 0.01 mM 1a, 25-
dihydroxycholecalciferol (Sigma Aldrich, D1530). Cells were cultured at a
density of 5000 cells/cm
2
. Fifty percent of fresh culture medium was
changed every 3rd day. hFOBs were used as positive control cells for the
production of DNA standards. Primary adipose tissue-derived progenitors
cells were used as study samples in monolayer (2D) and scaffold (3D)
scenarios. As a proof of principle, only samples induced after 28 days with
osteogenic culture media were chosen for subsequent assay with PCR.
2.2. Scaffold preparation and cell seeding
Medical-grade Polycaprolactone (PCL) scaffolds (Osteopore, Singa-
pore) with lay-down pattern angles of 01/601/1201 were used in the 3D
group. The surfaces of the scaffolds were treated in 5
M NaOH at 37 1C for
24 hours to enhance its hydrophilicity followed by extensive washing with
PBS. The scaffolds were cut into dimensions of 10 10 4 mm and
sterilized with 70% ethanol, washed with sterile PBS and dried. Passage 2
adipose-derived progenitor cells were seeded into the scaffolds with fibrin
glue (Baxter
s
Tisseel VH kit). Reconstituted thrombin (final concentra-
tion of 4 I.U./ml) in sterile 40 mM calcium chloride solution containing
500 000 cells (per scaffold) in DMEM was mixed with 2% fibrinogen in
sterile deionized water in situ. The fibrin-seeded scaffolds were placed in
culture plates and incubated for two hours at 37 1C/5% CO
2
prior to
media top-up.
2.3. Total RNA extraction
For the monolayer cultures, RNA was extracted with Trizol (Invitro-
gen, 15596—018). The aqueous portion was extracted with chloroform
and subsequent RNA was precipitated with 70% ethanol in diethylpyr-
ocarbonate treated water and cleaned up with RNeasy columns (Qiagen,
74106).
For the scaffold cultures, media was removed from the scaffold by
spinning at 1000g for 10 s. The scaffold was washed by dipping in sterile
PBS and spun again to remove excess liquid. Two more washes were
repeated. After the last spin, RLT lysis buffer (Qiagen, 74106) containing
b-mercaptoethanol was added to the scaffold and incubated at room
temperature for 15 min. The RLT buffer was collected with spinning and
fresh RLT buffer was added into the scaffold and incubated for 5 min and
ARTICLE IN PRESS
D.T. Leong et al. / Biomaterials 28 (2007) 203–210204
Page 2
collected. RLT buffer containing RNA was then processed according to
the manufacturer’s protocol (Qiagen, 74106).
2.4. Deoxyribonuclease I (DNase I) treatment and reverse
transcription (RT)
Before RT, the RNA samples were subjected to DNase I (Fermentas,
#EN0521) treatment to remove genomic DNA carryover to the RT
reaction. DNA digestion was carried out at 37 1C for 30 min. The samples
were then mixed with 70% ethanol, purified with RNeasy columns and
quantified. Purified RNA was reverse transcribed to cDNA using MuLV
reverse transcriptase RNase H
(Fermentas, #EP0451). cDNA of the two
study samples (adipose-derived progenitor cells) were diluted with sterile
deionized water to 10 ng/ml (2D_10 ng and 3D_10 ng). The samples were
then serially diluted by 10 and 100 times to represent starting total RNA
levels of 1 ng/ml (2D_1 ng and 3D_1 ng) and 0.1 ng/ml (2D_0.1 ng and
3D_0.1 ng), respectively.
2.5. Primer design and standards preparation
Primers were designed and found unique for osteocalcin and
b-actin after ‘‘blastn’’ (http://www.ncbi.nlm.nih.gov/BLAST/). Primer
sequences: osteocalcin—GCAGAGTCCAGCAAAGGT; CAGCCATTG
ATACAGGTAGC. b-actin—TGTGGCATCCACGAAACTAC; GGA
GCAATGATCTTGATCTTCA. PCR was run to amplify the cDNA
product from the hFOBs samples. The single amplified PCR product
was verified based on size in a 3% agarose gel under UV illumination.
The gel band containing the DNA target was excised and digested
to recover and purify the amplified product (Fermentas DNA extraction
kit, #K0513). The specificity and the uniqueness of the products from
each primer was further confirmed by sequencing of the product
(Research Biolabs, Singapore). The concentration of the amplified
product was measured with a spectrophotometer (Nanodrop ND-1000).
Using the average molecular weight of the product and Avogadro’s
constant, the number of copies per unit volume was calculated. The
volume of the purified linear dsDNA standards was adjusted to 10
10
copies per ml. This stock solution was serially diluted to obtain a standard
series from 10
9
to 10 copies per ml with each step differing by 10 fold.
When assaying the samples for osteocalcin and b-actin, the corresponding
standards series was run under the same conditions and the copy numbers
of samples was determined by reading off the standards series with the C
t
values of the samples.
2.6. Absolute quantitative real-time PCR (aqPCR)
After assuring the suitability of the primers for their uniqueness to
amplify a single PCR product (validated with sequencing results),
quantitative real-time PCR was run on samples and standards (n ¼ 3
per sample or standard) using a thermocycler (Stratagene Mx 3000 P) and
SYBR Green PCR kit (Quantitect Qiagen, 204143). PCR+dissociation
routine: 95 1C for 10 min, 45 cycles of 94 1C30s,601C45s,721C30s,
95 1C 1 s, 60 1C 30 s, slow ramp up to 95 1C at 0.5 1C per second
with continuous measurement, 95 1C10s,251C 30 s, end. (Dissociation
phase in italics.)
2.7. Gel electrophoresis and analysis of samples and standards
Samples reactions (n ¼ 3 per sample) from the above-described real-
time PCR protocol was separated in a 3% agarose gel and visualized
under UV illumination and digital images were captured. The band images
were analyzed with Bio-Rad Quantity One version 4.6.1 software.
3. Results
3.1. End-point PCR gel electrophoresis
Since b-actin and osteocalcin could be detected in end-
point PCR and later in aqPCR, mRNA could be
sufficiently isolated and reverse transcribed to cDNA for
quantitative analysis with PCR methods (Fig. 1). Gel
electrophoresis showed clear single bands at the expected
sizes of 182 bp b-actin (Fig. 1) and 71 bp for osteocalcin
(Fig. 1) showing specific amplicons. There was decreasing
band intensity from 10 to 0.1 ng for both environments and
genes. Non -template control (NTC) did not show any
bands.
3.2. Standards plots for b-actin and osteocalcin
For illustration purposes, Fig. 2a and b only one
standard and one sample was presented. The amplification
plots showed the standard and sampl e intercepting the
threshold line at two-thr eshold cycle number (C
t
) but at the
exponential phase of the PCR (Fig. 2a and b). Both plots
for b-actin however reached similar plateau levels at the
end of the PCR (Fig. 2a dotted lined box). The dissociation
curves of osteocalcin and b-actin for the samples and
standards showed only one peak (Fig. 2c and d). The T
m
for b-actin and osteocalcin were found to be 83.2 and
85.1 1C, respectively (Fig. 2c and d). The no-template–con-
trol dissociation curve showed a flat profile. C
t
values for
standards values ranging from 10 to 10
9
copy numbers of
ARTICLE IN PRESS
10 1 0.1 10 1 0.1 L 10 1 0.1 10 1 0.1 NTC OC NTC BA
Osteocalcin
β-actin
monolayer
scaffold monolayer
scaffold
Fig. 1. Gel electrophoresis of standards and samples. Visual inspection revealed a decrease in band intensity from 10 to 0.1 ng groups. NTC—no template
control; L—100 bp DNA ladder, lowest band 100 bp.
D.T. Leong et al. / Biomaterials 28 (2007) 203–210 205
Page 3
b-actin and osteocalcin fell along a straight semi-log
trendline with a R
2
value of 0.9902 and 0.9964, respectively
(Fig. 2e). The gradient of the standards plot for b-actin and
for osteocalcin was 3.4399 and 3.6721, respectively
(Fig. 2e). From these gradients, the PCR efficiency of
b-actin and of osteocalcin was calculated to be 95.39% and
87.2%, respectively according to the theoretical efficiency
equation: Efficiency ¼ (10
1/gradient
1) 100%.
ARTICLE IN PRESS
Standard curves for beta actin and osteocalcin
y = -3.4399x + 43.596, Efficiency = 95.3%
R
2
= 0.9902
0
5
10
15
20
25
30
35
40
45
0246810
C
t
beta actin
osteocalcin
log
10
(copy numbers)
Amplification Plots for osteocalcin
-0.03
0.07
0.17
0.27
0.37
0.47
0.57
1 3 5 7 9 111315171921232527293133353739414345
Cycles
Fluorescence (dRn)
Amplification Plots for beta-actin
-0.04
0.06
0.16
0.26
0.36
0.46
0.56
0.66
1 3 5 7 9 1113 15171921232527293133353739414345
Cycles C
t
10
5
standard
3D_1ng sam
p
le
Dissociation Curve for osteocalcin
-70
30
130
230
330
430
530
630
730
830
930
1030
1130
1230
1330
60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90 92 94
Temperature (°C)
Fluorescence (-R' (T))
Dissociation Curve for beta-actin
-48
52
152
252
352
452
552
652
752
852
60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90 92 94
Temperature (°C)
Fluorescence (-R'(T))
T
m
10
5
standar
d
2D_10ng sample
Fluorescence (dRn)
y = -3.6721x + 40.61; Efficency = 87.2%
R
2
= 0.9964
(b)
(a)
(c)
(d)
(e)
Fig. 2. Various plots for real-time PCR run of b-actin. (a) Amplification plots for the 10
5
standard and the 3D_1 ng sample. NTC reaction is flat showing
no templates present in the reaction mix. C
t
—threshold cycle where amplification plot cuts the threshold level. At the end of the entire PCR reaction, both
standard and sample reactions reached approximately the same level of fluorescence (dotted box). (b) Dissociation curve (dR/dT vs. T)(R ¼ normalized
fluoresence; T ¼ temperature) shows single dissociation peak for the 10
5
standard and the 3D_1 ng sample. Single PCR amplicon has a melting
temperature of 83.2 1C and showed unique PCR product. (c) Similar graphs obtained for the osteocalcin real-time PCR. (d) T
m
for this case was 85.1. (e)
Standards (10–10
9
) plot for b-actin and osteocalcin from C
t
values of standards. R
2
value of 0.9902 and 0.9964 for b-actin and osteocalcin. Theoretical
PCR efficiency was 95.3% and 87.2% for b-actin and osteocalcin, respectively based on the theoretical equation: Efficiency ¼ (10
1/gradient
1) 100%.
D.T. Leong et al. / Biomaterials 28 (2007) 203–210206
Page 4
3.3. aqPCR for b-actin and osteocalcin
Fig. 3 summarized the number of copies obtained per
monolayer and scaffold sample and their appropriate
10- and 100-fold dilutions. It was observed that the
monolayer express ed 51.2- and 17.4-fold higher levels of
b-actin and osteocalcin, respectively than the scaffold
culture. To obtain a relative quantification, after normal-
izing the osteocalcin counts by b-actin counts, the scaffold
culture instead expressed 2.94-fold higher osteocalcin than
that of monolayer.
3.4. aqPCR were able to accurately detect 10- and 100-fold
dilution in the samples
The data when assayed with gel electrophoresis and
aqPCR were plotted in Fig. 4 for both b-actin and osteo-
calcin. There was a decreased ratio of the 1 and 0.1 ng
groups when normalized to the 10 ng groups for both
samples and genes. However, the decrease in ratio when
assayed wi th end-point PCR was less drastic than expected
(Fig. 4). Conversely, aqPCR were able to detect the fold
changes accurately (Fig. 4).
ARTICLE IN PRESS
0
5000
10000
15000
2D_10ng_OC 2D_1ng_OC 2D_0.1ng_OC
Group
0
500
1000
3D_10ng_OC 3D_1ng_OC 3D_0.1ng_OC
Group
Copies of beta-actin in the 2D group
determined by aqPCR
0
5000000
10000000
15000000
20000000
2D_10ng_BA 2D_1ng_BA 2D_0.1ng_BA
Group
Copies of beta-actin in the 3D group
determined by aqPCR
0
100000
200000
300000
400000
3D_10ng_BA 3D_1ng_BA 3D_0.1ng_BA
Group
1.34×10
4
1.59×10
3
1.19×10
2
7.72×10
2
6.61×10
1
6.57
1.45×10
7
1.44×10
6
1.39×10
5
2.83×10
5
2.64×10
4
2.59×10
3
Copies of osteocalcin in the 2D group
determined by aqPCR
Copies of osteocalcin in the 3D group
determined by aqPCR
CopiesCopies
CopiesCopies
Fig. 3. Summary of data obtained from aqPCR. The four graphs presented the actual copy numbers of each sample for osteocalcin and b-actin genes.
0
0.2
0.4
0.6
0.8
1
1.2
2D 1ng
OC
2D 0.1ng
OC
3D 1ng
OC
3D 0.1ng
OC
2D 1ng
BA
2D 0.1ng
BA
3D 1ng
BA
3D 0.1ng
BA
samples
Gel EP
Real time
expected
normalized ratio
(to corresponding 10ng level)
Graph of normalized ratio of serially diluted samples using gel electrophoresis
and absolute quantitative real time PCR
Fig. 4. Summary of data obtained from gel electrophoresis and aqPCR. Normalization of the readings of the gel intensity with the reading of the
corresponding 10 ng sample for each group. Image analysis of end-point PCR however, did not show the 10-fold differences expected from the serially
diluted samples for both genes of monolayer and scaffold groups. Conversely data from aqPCR showed the 10-fold differences. The expected ratio column
was added for better comparative visualization. The expected ratio column is 0.1 for the 1 ng groups and 0.01 for the 0.1 ng groups.
D.T. Leong et al. / Biomaterials 28 (2007) 203–210 207
Page 5
3.5. Effects of DNase I treatment
Referring to Groups 1 and 2, the difference was the
additional peak at 86.8 1C for Group 2 (Fig. 5). This
additional peak of Group 2 coincided with that of Group
4, which had no reverse transcription react ion (Fig. 5).
Comparing Groups 3 and 4, there was significant gDNA
contamination and our DNA digestion treatment was
sufficient in removing that contam ination (Fig. 5). The
NTC group data showed there was no inherent DNA
contamination in the reagents.
4. Discussions
PCR, together with the discovery of the RNA-dependent
DNA polymerase (reverse transcriptase, RT) has allowed
us to study gene expression at a molecular level. Through
the amplification of the signal with PCR, it is now possible
to detect extremely low levels of mRNA. Of interest to the
biomaterials field [13,14], this technology was applied to
detect the expression of a particular gene of interest.
End-point PCR is commonl y used because it is
inexpensive and does not require specialized equipment.
However, end-point PCR could and should only be used to
detect the presence or absence and not to quantify gene
expression levels. In our study, the conclusions about the
level of PCR products obtained from real-time and end-
point PCR are disparate. It was not surpri sing since end-
point PCR measured the amount of products at the end of
the PCR. By taking readings at the exponential phase of
the PCR, aqPCR method could quantify the starting
level of cDNA of each gene to a level of at least in the
order of sub-1000 copies of osteocalcin per 10 ng of starting
total RNA.
Relative quantifica tion real-time PCR is a big leap in
quantifying gene expression compared to end-point PCR,
it still has its insufficiencies. Relative quantification
concluded in a perceived higher expression of osteocalcin
in the scaffold culture compared to the monolayer culture
whilst aqPCR concluded otherwise. In this study, though
the relative quantification of osteocalcin was not per-
formed conventionally, it still illustrated the skewing effects
ARTICLE IN PRESS
Group
12 34
Reverse Transcription + +
−−
Genomic DNA digestion + +
NTC
3
1
2
4
Group 1
Legends
Group 3
Group 2
Group 4
NTC
800
700
600
500
400
300
200
100
0
60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90 92 94
Temperature (°C)
Fluorescence (-R' (T))
Dissociation Curve
Fig. 5. Effect of DNase treatment on real-time PCR interpretation. After DNase treatment and RT, dissociation curve of Group 1 showed only one peak
at 83 1C. Groups 2 and 4 showed the presence of gDNA which resulted in an additional amplicon of T
m
of 87 1C. This additional amplicon was not present
after gDNA digestion (Groups 1 and 3). Group 3, after DNase treatment but no RT, did not have any peak. This meant that the peak observed in Group 1
was due to the mRNA that was transcribed to cDNA and then underwent PCR, resulting in a peak at 83 1C in the dissociation curve. However, after RT
but no gDNA digestion (Group 2), the additional amplicon was still present (box). NTC showed no amplification showing that the DNA contamination
was not from the real-time PCR reaction mix but from the samples themselves.
D.T. Leong et al. / Biomaterials 28 (2007) 203–210208
Page 6
of the normalizing gene (b-actin). aqPCR can be more
robust than relative quantification because aqPCR is
independent of any normalizing gene. In aqPCR, the only
normalization done between samples was the use of
starting amount of total RNA that represented the entire
transcriptome, thus decreasing the error. Conversely,
relative quantification can be biased by just one normal-
ization gene. Relative quantification is commonly
calculated with the 2
DDCt
formulae. The assumption for
valid use of this formulae is that the PCR efficiency of the
gene of interest and the normalizing gene must be similar
(5% difference) and close to 100%. These requirements
are not easily met. Common normalizing genes like
housekeeping genes may not be consistently expressed in
all the samples [15]. It is possible to use two or more
normalizing genes but it will complicate analyses because it
is not expected that the housekeeping genes are expressed
in equal weightage. With more normalizing genes, the runs
will also require more sample cDNA. Another difficulty is
to design the PCR reactions of both the gene of interest
and normal izing gene reactions to be similar and close to
100% efficiency.
The gold standard in gene expression quantification
is not to quantify relative to normalizing genes but
to ascertain actual copy numbers of transcripts in a
given mass of sample. The conventional absolute
quantification real-time PCR required molecular cloning
steps to produce the standards. We have previously
used SYBR Green I to quantify highly homologous
alternatively spliced sequences without the need of
probes using linearized plasmids encoding target
sequences [16]. In addition, the use of stem-loop mediated
RT real-time PCR showed a high degree of sensitive and
specificity for the quantification of replicative viruses [17].
In this study, we showed that even without molecular
cloning technqiues, a standard curve of sufficient accuracy
and linear range could be produced. When using conven-
tional aqPCR with molecular cloning steps, Whelan et al.
showed good linearity of R
2
¼ 0:9956 and a range of
8–1.6 10
7
plasmid molecules [13]. Our aqPCR data
showed similar ranges (10
1
–10
9
copies) with similar
linearity (R
2
¼ 0:9964 for osteocalcin) albeit a difference
in genes compared but none theless this showed that our
aqPCR method is viable.
Our method however needs to take into account the
following points. It was assumed that the efficiencies of RT
when producing the standar ds and the cDNA of the
samples were the same. In this study (as with other studies),
different cell types were used to produce the standards and
samples cDNA. If one is using a different cell type, it is
possible that the transcriptome profile of the positive
control cells (used as templates for standard s creation) that
expresses the gene of interest is very different from that of
the sample, thereby resulting in differences in RT
efficiency. Another issue was that the PCR of the standards
took place in an environment where the only templates
present were targets. This is not true for the samples since
the cDNA population is almost an entire replication of the
transcriptome, comprising of the target template (if any)
and many other genes. Therefore, the PCR could be
different in the standards reaction and the samples
reaction. However, the latter issue was also relevant in
conventional aqPCR where cloning steps were necessary.
Nonetheless, our method gave good resolution and offered
quantitative advantage over the relative quantification
method and end- point PCR method.
In the real-time PCR sectio n, we had used SYBR
Green I, a DNA intercalating fluorescent dye. This
chemistry is the simplest and least expensive of the
presently known types of quantifying PCR progress [18].
Using SYBR Green I chemistry, our aqPCR technology
could be easily adopted by biomat erials and tissue-
engineering laboratories with existing protocols for
absolute quantitative real-time PCR. SYBR Green I
obviated the need for target-specific fluorescent probes
and had the advantage of using the same PCR master
mix for many genes of interest. However, its universality is
also its bane as its specificity is determined entirely by its
primers and the PCR conditions. PCR products
arising from specific and non-specific PCR are similarly
detected with the SYBR Green I method. This could result
in false positives. Therefore, SYBR Green assays required
careful optimization of the PCR conditions and design of
the primers. To further differentiate specific from non-
specific PCR products, dissociation curve analysis is
usually carried out.
Another confounding problem with using SYBR Green I
chemistry is the contamination from genomic DNA
(gDNA). As gDNA will also contain the gene sequence,
it is likely that the gene of interest sequence in gDNA will
be co-amplified with that of cDNA. This co-amplification
results in false positive values for end-point PCR or in real-
time PCR methods where SYBR Green I is used. If there is
gDNA contamination, there will be an extra PCR product
that can be detected in the dissociation curve as another
peak with a higher T
m
. The quantification after repurifica-
tion was to ensure that there were similar levels of input
RNA for the PCR reaction comparison across samples. In
this study, we showed that gDNA could be detected in the
dissociation curve and illustrated the importance of
including a DNA digestion step prior to reverse transcrip-
tion (Fig. 5).
5. Conclusion
In this study, absolute quantification with PCR stan-
dards in a real-time PCR protocol gave better accuracy
than using end-point PCR methodology. By using this
method of aqPCR, there is no need for incorporating
molecular cloning steps into the existing real-time
PCR protocol to quantify gene expression. As shown,
real-time PCR users should be aware of the possibility of
genomic DNA contamination in any total RNA extraction
protocols.
ARTICLE IN PRESS
D.T. Leong et al. / Biomaterials 28 (2007) 203–210 209
Page 7
References
[1] Honda MJ, Yada T, Ueda M, Kimata K. Cartilage formation by
serial passaged cultured chondrocytes in a new scaffold: hybrid 75:25
poly(
L-lactide-epsilon-caprolactone) sponge. J Oral Maxillofac Surg
2004;62:1510–6.
[2] Yamane S, Iwasaki N, Majima T, Funakoshi T, Masuko T, Harada
K, et al. Feasibility of chitosan-based hyaluronic acid hybrid
biomaterial for a novel scaffold in cartilage tissue engineering.
Biomaterials 2005;26:611–9.
[3] Klapperich CM, Bertozzi CR. Global gene expression of cells
attached to a tissue engineering scaffold. Biomaterials 2004;25:
5631–41.
[4] van Susante JLC, Pieper J, Buma P, van Kuppevelt TH, van
Beuningen H, van Der Kraan PM, et al. Linkage of chondroitin-
sulfate to type I collagen scaffolds stimulates the bioactivity of seeded
chondrocytes in vitro. Biomaterials 2001;22:2359–69.
[5] Deng Y, Lin XS, Zheng Z, Deng JG, Chen JC, Ma H, et al.
Poly(hydroxybutyrate-co-hydroxyhexanoate) promoted production
of extracellular matrix of articular cartilage chondrocytes in vitro.
Biomaterials 2003;24:4273–81.
[6] Rappolee DA, Mark D, Banda MJ, Werb Z. Wound macrophages
express TGF-alpha and other growth factors in vivo: analysis by
mRNA phenotyping. Science 1988;241:708–12.
[7] Wang T, Brown MJ. mRNA quantification by real time TaqMan
polymerase chain reaction: validation and comparison with RNase
protection. Anal Biochem 1999;269:198–201.
[8] Niesters HG. Quantitation of viral load using real-time amplification
techniques. Methods 2001;25:419–29.
[9] Mauney JR, Jaquiery C, Volloch V, Heberer M, Martin I, Kaplan
DL. In vitro and in vivo evaluation of differentially demineralized
cancellous bone scaffolds combined with human bone marrow
stromal cells for tissue engineering. Biomaterials 2005;26:3173–85.
[10] Chen G, Sato T, Ushida T, Hirochika R, Tateishi T. Redif-
ferentiation of dedifferentiated bovine chondrocytes when
cultured in vitro in a PLGA–collagen hybrid mesh. FEBS Lett
2003;542:95–9.
[11] Giulietti A, Overbergh L, Valckx D, Decallonne B, Bouillon R,
Mathieu C. An overview of real-time quantitative PCR: applications
to quantify cytokine gene expression. Methods 2001;25:386–401.
[12] Whelan JA, Russell NB, Whelan MA. A method for the absolute
quantification of cDNA using real-time PCR. J Immunol Methods
2003;278(1–2):261–9.
[13] Vinatier C, Magne D, Weiss P, Trojani C, Rochet N, Carle GF,
et al. A silanized hydroxypropyl methylcellulose hydrogel for the
three-dimensional culture of chondrocytes. Biomaterials 2005;26:
6643–51.
[14] Lisignoli G, Zini N, Remiddi G, Piacentini A, Puggioli A, Trimarchi
C, et al. Basic fibroblast growth factor enhances in vitro mineraliza-
tion of rat bone marrow stromal cells grown on non-woven
hyaluronic acid based polymer scaffold. Biomaterials 2001;22:
2095–105.
[15] Bustin SA. Absolute quantification of mRNA using real-time reverse
transcription polymerase chain reaction assays. J Mol Endocrinol
2000;25:169–93.
[16] Yoong LF, Peng ZN, Wan G, Too HP. Tissue expression of
alternatively spliced GFRa1, NCAM and RET isoforms and the
distinct functional consequence of ligand induced activation of
GFRa1 isoforms. Brain Research Brain Res Mol Brain Res
2005;139:1–12.
[17] Anwar A, August JT, Too HP. A stem-loop-mediated reverse
transcription real-time PCR for the selective detection and quanti-
fication of the replicative strand of an RNA virus. Anal Biochem
2006;352:120–8.
[18] Klein D. Quantification using real-time PCR technology: applications
and limitations. Trends Mol Med 2002;8:257–60.
ARTICLE IN PRESS
D.T. Leong et al. / Biomaterials 28 (2007) 203–210210
Page 8
  • Source
    • "There was no significant difference (p > 0.05) with respect to viral load between ALT, AST and GGT enzymes, cytokine expression levels, and total and indirect bilirubin ( table 2) . The frequencies fit the criteria set out by Manly [26] in 2005. The genotype distribution in our study did not significantly differ from the distribution expected according to the Hardy-Weinberg equilibrium for the three genes analyzed (p > 0.05). "
    Full-text · Article · Apr 2016 · Intervirology
  • Source
    • "Absolute quantification of Spag11c mRNA levels was based on Leong et al. (2007). Briefly, cDNA samples (250 ng) from adult caput epididymis were amplified by PCR (40 cycles) in a final volume of 20 μl containing 20 mM Tris–HCl pH 8.4, 50 mM KCl, 1.5 mM MgCl2, 0.2 mM dNTPs, 2 U Taq polymerase (Invitrogen) and 0.4 μM of each Spag11c forward and reverse primers (Table 1). "
    [Show abstract] [Hide abstract] ABSTRACT: Herein, we characterized the spatio-temporal expression, cellular distribution and regulation by androgens of the β-defensin SPAG11C, the rat ortholog of the human SPAG11B isoform C, in the developing epididymis by using RT-PCR, in situ hybridization and immunohistochemistry. We observed that Spag11c mRNA was ubiquitously expressed in rat fetuses, but preferentially detected in male reproductive tissues at adulthood. SPAG11C (mRNA and protein) was prenatally mainly detected in the mesenchyme of the Wolffian duct, switching gradually after birth to a predominant localization in the epididymis epithelium during postnatal development. In the adult epididymis, smooth muscle and interstitial cells were also identified as sources of SPAG11C. Furthermore, SPAG11C was differentially immunolocalized on spermatozoa surface during their transit from testis throughout caput and cauda epididymis. Developmental and surgical castration studies suggested that androgens contribute to the epididymal cell type- and region-specific modulation of SPAG11C mRNA levels and immunolocalization. Together our findings provide novel insights into the potential role of β-defensins in the epididymis. Copyright © 2015. Published by Elsevier Ireland Ltd.
    Full-text · Article · Feb 2015 · Molecular and Cellular Endocrinology
  • Source
    • "However, release of acidic degradation products can cause a severe inflammatory response in the body (Bergsma et al. 1993; Tam et al. 1996; Martin et al. 1996; Suuronen et al. 1998; Tatakis and Trombelli 1999; Bostman and Pihlajamaki 2000). Since the 1990s, other types of aliphatic polyester: polyhydroxyalkanoates (PHA) particularly poly-3- hydroxybutyrate (P3HB), copolymer of 3-hydroxybutyrate and 3-hydroxyvalerate (PHBV), poly-4-hydroxybutyrate (P4HB), copolymers of 3-hydroxybutyrate and 3-hydroxy- hexanoate (PHBHHx) and poly-3-hydroxyoctanoate (Leong et al. 2007 ) have been increasingly investigated as scaffolding materials for tissue engineering application due to their high biocompatibility (Chen and Wu 2005; Misra et al. 2006). They are natural thermoplastic polyesters produced by a wide variety of microorganisms under imbalanced growth conditions (Doi et al. 1995; Li et al. 2005). "
    [Show abstract] [Hide abstract] ABSTRACT: Tissue engineering is essentially a technique for imitating nature. Natural tissues consist of three components: cells, signalling systems (e.g. growth factors) and extracellular matrix (ECM). The ECM forms a scaffold for its cells. Hence, the engineered tissue construct is an artificial scaffold populated with living cells and signalling molecules. A huge effort has been invested in bone tissue engineering, in which a highly porous scaffold plays a critical role in guiding bone and vascular tissue growth and regeneration in three dimensions. In the last two decades, numerous scaffolding techniques have been developed to fabricate highly interconnective, porous scaffolds for bone tissue engineering applications. This review provides an update on the progress of foaming technology of biomaterials, with a special attention being focused on computer-aided manufacturing (Andrade et al. 2002) techniques. This article starts with a brief introduction of tissue engineering (Bone tissue engineering and scaffolds) and scaffolding materials (Biomaterials used in bone tissue engineering). After a brief reviews on conventional scaffolding techniques (Conventional scaffolding techniques), a number of CAM techniques are reviewed in great detail. For each technique, the structure and mechanical integrity of fabricated scaffolds are discussed in detail. Finally, the advantaged and disadvantage of these techniques are compared (Comparison of scaffolding techniques) and summarised (Summary).
    Full-text · Article · Dec 2014
Show more