Whitepaper serumplasma

Scientific article

miRNA profiling from blood — challenges
and recommendations
Jonathan Shaffer,1 Martin Schlumpberger,2 and Eric Lader1
QIAGEN Inc., 6951 Executive Way, Frederick, MD, USA;

1

QIAGEN GmbH, QIAGEN Strasse 1, Hilden, Germany.

2

Abstract: The discovery of stable miRNA species circulating in blood has led to increased research
focus on disease-related variations in serum and plasma miRNA expression and the possibility
that such variations could serve as noninvasive biomarkers for disease. Working with serum and
plasma miRNA presents various challenges in purification and characterization. In this paper, we
outline QIAGEN® recommendations for robust purification and quantification, as well as reliable
data normalization and analysis.

Introduction
In a few short years, microRNA (miRNA) research has advanced

stabilization, and purification solutions, as well as a robust

from a single publication reporting the discovery of a small,

miRNA quantification system that enables efficient profiling of

regulatory, noncoding RNA in Caenorhabditis elegans to

miRNA changes in serum, plasma, and many other sample types.

thousands of publications describing both the mechanism behind

RNA interference (RNAi) and the many critical roles that miRNAs
play in the regulation of gene expression. In humans, the annotated
miRNome has rapidly expanded from only a handful of sequences
to nearly 2000 in miRBase version 18 (www.mirbase.org).
Large-scale cloning and next-generation sequencing studies have
revealed surprising diversity in the mature miRNA population.
Heterogeneity is generated by variations in the processing of
precursor miRNA (1), as well as by posttranscriptional RNA
editing (2). miRNAs have been demonstrated to act synergistically
on target mRNAs, further increasing their potential regulatory

Contents
Introduction�� 1
Stable, circulating miRNA___________________________ 2
Challenges of miRNA purification from blood______________ 2
miRNA purification from fresh whole blood�� 3
miRNA purification from nucleated cells�� 3
miRNA purification from serum and plasma�� 4
miRNA profiling by real-time RT-PCR______________________ 4

repertoire. Considering the diversity of mature miRNAs found

Reverse transcription from serum and plasma samples__ 4

in the cell and the potential for cooperative regulation of target

Real-time PCR of serum and plasma samples__________ 5

genes, the potential complexity of miRNA effects on cellular

Quality control of serum and plasma miRNA_______________ 6

functions is vast.

Exogenous controls for circulating miRNA profiling_________ 6

miRNA research has made a significant impact on all aspects

Data normalization for serum and plasma samples_________ 6

of biomedical research, from elucidating basic functions and

Examples of data normalization_____________________ 8

mapping pathways in model systems to understanding gene

Biomarker discovery recommendations____________________ 8

expression changes in cancer and creating new possibilities for

Conclusion____________________________________________ 9

molecular diagnostics and nucleic-acid–based drugs. QIAGEN

References�� 10

supports this research by providing optimized sample collection,

Ordering information�� 10

36
CT meanmean (plasma at 25°C for 24 h)
CT (plasma at 25°C for 24 h)

Stable, circulating miRNA
The first reports that miRNAs could be found at stable levels in
serum and plasma were initially met with skepticism, but were
quickly confirmed by others (3, 4, and Figure 1). This discovery
generated high interest in the possibility that changes in these
miRNA levels could be used as noninvasive biomarkers for a
variety of indications. Evidence is accumulating that these circulating miRNAs can indeed be used as biomarkers to identify
and monitor a variety of cancers and other diseases (Figures

36
32
32
28
28
24
24
20

R2 = 0.9379

20

R2 = 0.9379

16

2 and 3).

16

Circulating miRNA is presumably not naked miRNA, which would

16
16

be degraded within seconds due to the high levels of nucleases
in blood. Several reports have demonstrated that stability results
CT meanmean (plasma thawed 10x)
CT (plasma thawed 10x)

and specific proteins (5–7). Other studies have found miRNA
contained within circulating exosomes or other microvesicles (8).
While there is no longer any doubt that a stable population of
miRNAs exist in circulation, nothing is yet known about their
purpose and whether they play any role in normal physiology.
It is also unknown how and why disease causes changes in the
levels of specific circulating miRNAs. It is possible that cell lysis
or an increase in the number of exosomes shed from diseased cells

Exosomes
microvesicles

Ago
Ago-2–miRNA
complexes

HDL-mediated
miRNA transport

Other protective
protein

36
36

36
32
32
28
28
24
24
20

R2 = 0.9777

20

R2 = 0.9777

16
20
24
28
32
36
CT mean (plasma thawed 1x)
16
20
24
28
32
36
CT mean Fresh thawed 1x)
Figure 2. Stability of plasma miRNA. (plasma plasma samples or plasma samples
stored at 25°C for 24 hours were used for RNA purification using the miRNeasy
Serum/Plasma Kit. Profiling was performed using the Human Serum Plasma
miScript® miRNA PCR Array. The equivalent of 20 µl plasma was used for each
array. Pairwise comparison of the data was performed using a scatter plot of CT
values. Incubation for 24 hours at room temperature had only a minor effect on
expression levels. Storage of plasma at room temperature is not recommended.
However, these results show that miRNA in these samples is remarkably stable.
Plasma samples were either frozen once or subjected to 10 freeze–thaw
cycles prior to RNA purification using the miRNeasy Serum/Plasma Kit. Profiling
and data analysis was performed as described above. Multiple freeze–thaw
cycles had no effect on reproducibility of miRNA profiling from plasma.
16

miRNA

24
28
32
CT mean (fresh plasma)
20
24
28
32
CT mean (fresh plasma)

36

from the formation of complexes between circulating miRNA

contribute to increased levels of certain circulating miRNAs.

20

16

Challenges of miRNA purification from blood
Neutrophils

Eosinophils

Blood has always presented a unique challenge to RNA isolation

Basophils

technologies. Compared to cells or solid tissue samples, there are
relatively few nucleated cells in blood; therefore, larger sample
volumes must be processed. Many of the cells in blood are
Lymphocytes

Monocytes

Platelets

poised to rapidly respond to changes in their environment and,

Erythrocytes

as a result, show changes in gene expression almost immediately

Figure 1. Circulating miRNA in blood. Circulating miRNA may be protected from
degradation by a variety of mechanisms, including formation of complexes with
proteins or secretion within exosomes. Human blood consists of red blood cells
or erythrocytes, platelets, and white blood cells or leukocytes, which include
neutrophils, eosinophils, basophils, lymphocytes, and monocytes. These cells are
suspended in plasma.
2

after sample collection. The presence of high concentrations of
protein, including nucleases and other components, can interfere
with downstream enzymatic reactions. For example, heparin, a
common anticoagulant, can copurify with RNA and is a potent

www.qiagen.com

QIAGEN

miRNA purification from fresh whole blood

2-∆CT breast cancer serum

1.E+03

The optimal approach for miRNA purification from whole blood

1.E+02

is to collect fresh blood and process the sample as quickly as

1.E+01

possible. We recommend the PAXgene Blood miRNA Kit for
miR-34

human blood (with blood collected and stabilized in PAXgene

1.E+00

Blood RNA Tubes) or the RNeasy Protect Animal Blood Kit for

R2 = 0.9379

1.E-01

mouse, rat, and other small animal blood (with blood collected
and stabilized in RNAprotect Animal Blood Tubes). A single
3

2

E+
0
1.

1.

E+
0

1
1.

E+
0

0

1

E+
0
1.

E0
1.

1.

E0

2

1.E-02

2-∆CT normal serum

Figure 3. miRNA variation in normal and breast cancer serum. Total RNA was
isolated from normal (n = 10) and breast cancer (n = 3) serum samples. Profiling
was performed using the Human Serum Plasma miScript miRNA PCR Array.
A scatter plot of 2-ΔCT values demonstrates several significant differences between
mature miRNA expression levels of normal and breast cancer serum pools, including
hsa-miR-34a, which has been shown to be a blood-based marker for patients with
primary and metastatic breast cancer. Data sets were first calibrated using the
C. elegans miR-39 miScript Primer Assay which detects miRNeasy Serum/Plasma
Spike-In Control, and then normalized using commonly expressed miRNA targets.
Analysis of a larger panel of individual samples will be necessary to draw firm
conclusions about the expression of specific miRNAs.

PAXgene Blood RNA Tube enables collection of 2.5 ml blood,
while RNAprotect Animal Blood Tubes are available in 100 µl
or 500 µl capacities. Use of either of these tube types results in
disruption of all cellular content and stabilization of RNA prior
to sample preparation. Since cells are lysed, these methods do
not allow later separation of blood into cellular and fluid
components for miRNA analysis in these fractions. Ideally, fresh
blood should be used, since freezing and thawing blood prior
to processing can result in cell lysis and RNA degradation. RNA
preps from blood contain large quantities of globin mRNA;

inhibitor of enzymatic reactions. Even a high-quality RNA prep
from blood can contain contaminants that inhibit real-time RT-PCR
if too much sample is used in the reverse-transcription reaction.
QIAGEN sample handling and RNA purification solutions are
optimized depending on whether the downstream profiling
application requires all miRNA in whole blood, the miRNA
component of the nucleated cell fraction, or extracellular circulating
miRNA (Table 1).

however, this does not affect miRNA quantification by real-time
RT-PCR.
miRNA purification from nucleated cells
The simplest way to isolate miRNA from the cellular, non-red blood
cell (RBC) fraction of fresh whole blood is to first lyse RBCs by
addition of 1.5 volumes cold 1 mM EDTA and then pellet
the intact cells by centrifugation. Carefully remove the entire
supernatant, then follow the protocol provided in the miRNeasy

Table 1. Kits for purification of total RNA including miRNA

Micro Handbook, miRNeasy Mini Handbook, or miRNeasy 96
Handbook for cells, starting with cell pellets. If the blood was

Kit

Starting material

Benefit

previously frozen, it will no longer be possible to separate white

miRNeasy
Serum/Plasma Kit

Serum and plasma

Includes miRNeasy Serum/
Plasma Spike-In Control;
optimized for circulating,
cell-free miRNA

blood cells (WBCs) from other blood components because

miRNeasy
Mini Kit

Animal/human tissues,
cells

Optimized for cells and
tissues

miRNeasy
96 Kit

Animal/human tissues,
cells

High-throughput; optimized
for cells and tissues

of time may not provide a meaningful miRNA profile from the

miRNeasy
Micro Kit

Animal/human tissues
and cells

Optimized for low RNA
amounts and cells and
tissues

cells. If subfractionation of immune cells in blood by cell sorting

PAXgene® Blood
miRNA Kit

Human blood that
has been stabilized in
PAXgene Blood RNA
Tubes

Optimized for human
whole blood miRNA

Animal blood that
has been collected in
collected in RNAprotect ®
Animal Blood Tubes

Optimized for animal
whole blood miRNA

RNeasy Protect
Animal Blood Kit
®

Scientific article

freezing and thawing causes cells to burst. In this case, the only
option is to purify whole blood.
RNA purified from anticoagulated blood stored for any length
cellular component due to changes in gene expression in the
is performed during research, these purified cells can also be
treated as a cell pellet and processed using the miRNeasy
Micro, Mini, or 96 Kit.

www.qiagen.com

3

miRNA purification from serum and plasma

profiling, using a straightforward, optimized real-time RT-PCR

Plasma is a clear, yellow fluid that makes up about 55% of the

protocol (Figure 4). Patent-pending miRNA technology ingeniously

total blood volume. It contains fibrinogen and other clotting

integrates a universal tailing and reverse-transcription reaction

factors. Serum is blood plasma without fibrinogen or other clotting

with accurate expression analysis of distinct miRNA sequences

factors. Plasma is prepared by centrifugation of whole blood

that may only differ by a single nucleotide base. This technology

to remove RBCs and WBCs. Preparation of serum is more

allows the researcher to easily perform a comprehensive survey

complicated and involves allowing blood to clot, followed by

of relative miRNA expression in samples of interest. All miScript

centrifugation to remove the clot. We recommend using EDTA or

Primer Assays are designed to work under a common set of

citrate anticoagulated blood, and processing plasma rather than

cycling conditions and are thoroughly wet-lab verified to ensure

serum. This minimizes procedural variation caused by differences

high specificity and high sensitivity.

in clotting and subsequent collection of serum.
Since the mechanisms by which miRNAs enter circulation and
are protected from nucleases are still unknown, we currently
recommend profiling miRNA expression from total serum or
plasma rather than performing any enrichment of RNA–protein
complexes or exosomes. Purification of circulating miRNA from
serum or plasma can be performed using the miRNeasy Serum/
Plasma Kit. We recommend a starting volume of 100–200 µl serum
or plasma for RNA purification with the kit. Use of more than
200 µl serum or plasma per RNA prep is not recommended, as
the amount of contaminants present in larger volumes reduce
the binding capacity of the RNeasy MinElute® spin column and
increase the risk of copurification of inhibitors.
Purification and downstream real-time RT-PCR profiling work
equally well whether serum or plasma miRNA preps are used
as starting material. In our experience, profiling data is similar
from serum and plasma samples from the same donor. However
comparing a serum sample to a plasma sample from the same
donor shows more variability than comparison of 2 samples
of the same type (either serum or plasma). For this reason, we
recommend comparing the same sample types where possible
to minimize variability.

Reverse transcription from serum and plasma samples
Upstream of real-time PCR, the miScript PCR System includes an
integrated kit for reverse transcription: the miScript II RT Kit. The
kit comes with 2 different buffer chemistries, each optimized for
specific advantages. miScript HiSpec Buffer is a patent-pending
breakthrough chemistry that severely inhibits the activity of the
tailing reverse-transcription reaction on templates other than
miRNA-sized templates. This provides an exceptionally specific
cDNA synthesis reaction, virtually eliminating background from
longer RNA species. This advantage is not provided by any other
currently available chemistry, and is ideal for profiling experiments,
as melt curve and gel analysis become virtually unnecessary.
The limitation to this extraordinary chemistry is that miScript
HiSpec Buffer can only be used for mature miRNA quantification.
To measure pre-miRNA (with miScript Precursor Assays) or
mRNA (with QuantiTect® Primer Assays) from the same reversetranscription reaction, it is necessary to use miScript HiFlex Buffer
(also included in the miScript II RT Kit). Reverse transcription
with miScript HiFlex Buffer is an extremely efficient but nonbiased
reaction. With cDNAs prepared using miScript HiFlex Buffer, a
small percentage of miScript Primer Assays will show increased
background signal from cross reactivity with sequences from a
total RNA prep. These are easily distinguished by performing a

miRNA profiling by real-time RT-PCR
Real-time RT-PCR is a reliable, easily applicable technique for
miRNA quantification. Arrays of miRNA assays can be used to

melt curve analysis, as miScript Primer Assays melt in a specific
and very narrow temperature range.

profile hundreds of miRNAs to gain a snapshot of the expressed

We recommend a starting volume of 100–200 µl serum or

miRNAs in the cell at a certain timepoint, in a diseased or healthy

plasma for RNA purification using the miRNeasy Serum/Plasma

state, or after stress or drug treatment. Alternatively, individual

Kit. After elution, 1.5 µl of a 14 µl eluate should be used as a

miRNA assays can be used for experiments studying the expression

template for reverse transcription with the miScript II RT Kit. This

of a single miRNA or a small group of miRNAs. Individual miScript

provides enough cDNA for either one 384-well plate or four

Primer Assays and miScript miRNA PCR Arrays enable high-

96-well plates. Up to 10 µl RNA eluate has routinely been used

quality expression analysis from single miRNAs to miRNome-wide

as template without inhibiting the reverse-transcription reaction,

4

www.qiagen.com

QIAGEN

Blood collection
plasma/serum isolation

RNA prep

miRNeasy Serum/Plasma Kit
and miRNeasy Serum/Plasma
Spike-In Control

cDNA synthesis
miScript II RT Kit

Data analysis
normalization

Real-time PCR
miScript SYBR Green PCR Kit
+ miScript Primer Assays
or
miScript miRNA
PCR Arrays

PCR Array Data Analysis
Software

1.E+001

Extract

White blood cells
and platelets (1%)

AAAAAAA

Delta Rn

1.E-001
1.E-002
1.E-003

Red blood cells (45%)

0

4

8

12

16

20 24

28 32

36

40

Cycle number

AAAAAAA

1.E+001

TTT

10-3
10-2
10-1
-4

-3

-2

-1

23

01

4

1
10-1

1.E-000

TTT

10-4

Fold change ratio [log2]

Delta Rn

Bind and
wash

10-5

1

Normal breast

Plasma (55%)

p-Value for fold change

10-6

1.E-000

1.E-001
1.E-002
1.E-003

10-2
10-3
10-4
10-5

0

4

8

12

16

20 24

28 32

36 40

Cycle number

10-6

10-6

10-5

10-4

10-3

10-2

10-1

1

Breast tumor

Elute

Figure 4. miScript PCR System workflow.

which would provide enough cDNA for more than six 384-well
plates or two copies of the Human miRNome miScript miRNA
PCR Array.

Table 2. miScript miRNA PCR Array solutions for serum and
plasma samples
miScript miRNA PCR Array

Species

Complete miRNome

Human, mouse, rat, dog, rhesus macaque

Real-time PCR of serum and plasma samples

Serum Plasma

Human, mouse, rat

Once completed, a reverse-transcription reaction as described

Serum Plasma 384 HC

Human

above provides sufficient cDNA for real-time PCR profiling using
miScript miRNA PCR Arrays. Various array options are available

As an overview, profiling the miRNome with the miScript PCR

for profiling miRNA expression in serum or plasma samples

System involves setting up a reverse-transcription reaction,

(Table 2). Every array includes replicates of a miRNA reverse

performing reverse transcription for 1 hour, diluting the cDNA,

transcription control assay (miRTC) and a positive PCR control

adding master mix, and aliquoting into the miScript miRNA PCR

(PPC). These are quality control assays that can be used to

Array. After cycling under standardized conditions on any real-

determine the presence of reverse transcription and real-time

time PCR cycler, resultant data can be exported and uploaded

PCR inhibitors.

to the QIAGEN Website for analysis (Figure 5).

For quantification of smaller numbers of miRNAs, each reversetranscription reaction can be diluted to 220 µl with RNase-free
water and 1 µl can be used as template for single miRNA
assays, regardless of the final PCR reaction volume.
Scientific article

www.qiagen.com

5

1. Convert miRNA to cDNA in a one-step, single-tube
reverse transcription reaction.

Table 3. miRNAs commonly expressed in serum and plasma
samples*
hsa-let-7a

5'

miRNA

5'

5'

Single-tube

5'

First strand cDNA

TTT T TT

hsa-miR-423-5p

* isit the GeneGlobe® Web portal (www.qiagen.com/GeneGlobe) to order
V
miScript Primer Assays for detection of these miRNAs.

Exogenous controls for circulating
miRNA profiling

miScript Primer Assay
TTT T TT

cDNA

hsa-miR-222

3'

2. Combine cDNA with QuantiTect SYBR Green PCR
Mastermix, miScript Universal Primer, and water.
Aliquot mixture across miScript miRNA PCR Array.

5'

hsa-miR-191

hsa-miR-26b

Universal RT primer

hsa-miR-146a

hsa-miR-26a

Reverse
transcription

3'
reaction A A A A A(A)n
TTT T TT

hsa-miR-145

hsa-miR-25

3'
A A A A A(A)n

hsa-miR-126

hsa-miR-24

Poly(A) tail

hsa-miR-103a

hsa-miR-23a

Polyadenylation

3'

hsa-miR-93

hsa-miR-22

RNA sample 2

hsa-miR-92a

hsa-let-7c
hsa-miR-21

RNA sample 1

One of the challenges of miRNA profiling from serum or

3'

miScript Universal Primer

plasma is the lack of established housekeeping genes for data
normalization when using the ΔΔCT method of quantification.
Normalization strategies will be discussed in greater detail in the

qPCR using miScript
Primer Assay and
miScript Universal Primer

next section, but it is worth mentioning here that one frequently
used strategy for helping to identify and minimize variations in

3. Run in real-time PCR cycler.

circulating RNA recovery is to spike a synthetic RNA sequence
into the sample. In conjunction with the miRNeasy Serum/Plasma

4. Analyze data.

Kit, QIAGEN offers the miRNeasy Serum/Plasma Spike-In Control

1

Sample 2

10-1

that can be added into serum and plasma preparations during

10-2

the purification process after homogenization with QIAzol® Lysis

10-3

Reagent. This enables normalization for any nonspecific losses

10-4
10-5
10-6

10-6

10-5

10-4

10-3

10-2

10-1

incurred during miRNA purification. Even in the absence of specific

1

miRNA signals, the spike-in control should give a positive signal

Sample 1

Figure 5. miRNA profiling workflow.

with a reasonably constant CT value. The miScript Primer Assay

Quality control of serum and plasma miRNA

for this C. elegans miRNA target is included in the miRNeasy
Serum/Plasma Kit, and is also included on every miScript miRNA

The amount of RNA recovered from serum or plasma is typically

PCR Array. This control does not replace normalization with

too small to be quantified or qualified by traditional methods

invariant internal controls, but helps control for variations in

such as spectrophotometry readings with a Nanodrop or

recovery and amplification efficiency between RNA preps.

assessments with a Bioanalyzer. In addition, long RNA will
the miRNA contribution. However, several miRNA assays can

Data normalization for serum and
plasma samples

serve as controls, such as those listed in Table 3. These should

Small, noncoding RNAs, such as snoRNAs and snRNAs, are

be readily detectable in human serum or plasma samples,

typically used for data normalization in miRNA profiling

thereby serving as a check that the sample contains miRNA and

experiments in the same way that housekeeping genes are used

performs normally in real-time RT-PCR.

for normalization of mRNA expression data. Many of these small

likely be degraded, making it impossible to specifically measure

6

www.qiagen.com

QIAGEN

RNAs are widely expressed and usually show little variation in
different cell types. In reverse-transcription reactions using the

Methods for normalization of serum and plasma miRNA

miScript II RT Kit, these small RNAs are polyadenlyated and

expression data

reverse transcribed in the same way as miRNAs. As a result, they

■■ ormalization by plate mean: The plate mean is the
N

serve as controls for variability in sample loading and real-time
RT-PCR efficiency. However, these widely used controls are not
well expressed in human serum and plasma samples (Table 4).
In fact, if the mean CT of the snRNA/snoRNA miScript PCR
Controls is less than 32, this indicates that the samples have
0.1% cellular contamination. For this reason, it is necessary to
use alternative methods of normalization for serum and plasma
samples. As a first step, if the miScript Serum/Plasma Spike-In
Control (a synthetic C. elegans miR-39)

was added to the

homogenized serum or plasma samples prior to RNA purification,
CT values obtained using the C. elegans miR-39 miScript Primer
Assay can be used to calibrate the data sets being analyzed.
This calibration can resolve differences in recovery that may
occur during the purification procedure and in amplification
efficiency. Using the data presented in Figure 3, the mean
CT values for the C. elegans miR-39 miScript Primer Assay were
18.90 and 19.55 for normal serum and breast cancer serum
samples, respectively. For calibration, 0.65 (the difference
between the mean CT values) was added to every primer
assay in the normal serum sample. Depending on the type of
experiment and the number of ‘present calls’, several methods
of normalization are then available that yield acceptable results
(see box). Normalization by either plate mean or commonly
expressed miRNA targets are ideal methods for profiling

mean CT value of all the miRNA targets on the plate
(Figure 7A).

■■ ormalization using commonly expressed miRNA
N
targets: When profiling the miRNome for example,
many targets will not be expressed in the samples. In
this simple strategy, only the targets that are expressed
in all samples are used to calculate the mean value
(9). The greater the percentage of negatives on the
plate, the more accurate this approach is compared
to the plate mean approach (Figure 7B).

■■ ormalization using a panel of invariant miRNA: When
N
comparing 2 sample types, it is common that many
miRNAs are relatively invariant (Figure 6 and Figure
7C). These are easy to recognize if the samples are
of equal miRNA content and there was no significant
difference in real-time RT-PCR efficiency caused by poor
technique or contaminants. When sample input amount
is less well controlled, a software package such as
NormFinder (www.mdl.dk/publicationsnormfinder.htm)
can be used to analyze the data set to determine
which assays are the least variable and thus, the best
candidates for use as normalizers. NormFinder can
run as a template within Microsoft® Excel®.

SNORD95

32.92

31.33

SNORD96A

34.20

33.68

RNU6-2

36.08

33.31

Scientific article

-2
4
23
-5
p
m
iR
-9
2a

3a

-4

m

iR

-2

iR
m

m

iR

a

91

-1

46

iR

-1

iR

m

a

26

-1

34.78

iR

32.28

34.89

m

34.38

m

SNORD68
SNORD72

c

33.87

iR

35.47

20

m

SNORD61

22

a

CT mean: cancer serum

24

-7

CT mean: normal serum

26

le
t

miScript PCR Control

28

03

Table 4. miScript PCR Control expression in human serum samples

30

-1

experiments where single miScript Primer Assays are used.

32

t-7

miRNAs is ideal for Custom miScript miRNA PCR Arrays or

le

miRNA PCR Arrays. Normalization using a panel of invariant

CT mean (normal serum)

experiments using either miRNome or Pathway-Focused miScript

Figure 6. Relatively invariant circulating miRNAs in healthy volunteers. Raw CT
values are shown for 10 miRNAs whose expression levels vary less than 2-fold in
plasma between 9 normal serum samples, each from a different donor. Expression
profiling was performed using the miScript PCR System and the Human Serum
Plasma miScript miRNA PCR Array. These data suggest that, for a given cohort of
samples, suitable miRNAs can be found that can serve as endogenous normalizers.

www.qiagen.com

7

Once normalization values for the dataset have been obtained,

This method averages out individual variances and finds robust

a classic ΔΔCT calculation and a log2 transformation provide

differences between the populations. Next, the miRNAs of inter-

normalized fold-difference values for the miRNA targets. If

est are organized in a Custom miScript miRNA PCR Array and

technical replicates have been included, the variance for each

screened with every individual sample in each population.

sample/assay combination can be calculated. The miScript
PCR System and miScript miRNA PCR Arrays are extremely

12
12
12

robust, and array-to-array variability is negligible. For this reason,

8
8
8

rather than simple technical replicates of the PCR array, we
Fold regulation
Fold regulation
Fold regulation

recommend biological replicates, for example, increasing the
number of each sample type to be analyzed.
Examples of data normalization

4
4
4
0
0
0
-4
-4
-4

In the following example (Figure 7), we take the same data set
derived from profiling 3 normal serum and 3 breast cancer serum

-8
-8
-8

samples using the Human Serum Plasma miScript miRNA PCR
Array and apply each of the normalization methods described

12
12
12

above.
In general, if technical variations in sample handling, sample

8
8
8
Fold regulation
Fold regulation
Fold regulation

prep, and real-time RT-PCR set up are kept to a minimum, only
small differences should be expected from these different data
normalization methodologies. Targets that differ by less than 2-fold
would be expected to show the most impact from changes in
normalization methodology, while targets with large differences

4
4
4
0
0
0
-4
-4
-4

in expression level would be least affected. To be valuable as

-8
-8
-8

a biomarker, any correlation between circulating miRNA level
and disease would need to be robust enough to withstand the
normal variations of collection and sample prep and hold true

12
12
12

in a statistically significant population.

8
8
8
Fold regulation
Fold regulation
Fold regulation

Biomarker discovery recommendations
Serum and plasma samples provide a unique opportunity for
biomarker discovery. As the number of annotated mature miRNAs
continues to grow, it becomes increasingly important to optimize

the hundreds or even thousands. We have successfully used
the procedure outlined below to identify key serum or plasma
miRNAs associated with disease (see box). With clearly stratified
populations (for example, serum from normal donors and serum
from donors exclusively with breast cancer), we first screen 3
pools of 10 samples of each clearly defined population using
either the Human miRNome miScript miRNA PCR Array or

0
0
0
-4
-4
-4

miRNA biomarker screening methods, as statistical power
requirements can push experimental sample requirements into

4
4
4

-8
-8
-8
Figure 7. Data normalization strategies. Mature miRNA expression was profiled
in 3 normal serum and 3 breast cancer serum samples using the Human Serum
Plasma miScript miRNA PCR Array. Fold-changes for each miRNA (breast cancer
versus normal serum) were calculated with CT values normalized as follows:
CT values normalized to whole plate CT mean,
CT values normalized to
the plate CT mean of commonly expressed miRNAs (those miRNAs with CT values
35 in both sample sets),
CT values normalized to the CT mean of at least
4 miRNAs with little CT variation. In these experiments, hsa-miR-23a, hsa-miR-24,
hsa-miR-92a, and hsa-miR-126 exhibited a CT variation less than 1 between the
samples being compared.

Human Serum Plasma 384HC miScript miRNA PCR Array.

8

www.qiagen.com

QIAGEN

To demonstrate the robustness of this approach: instead of
profiling the human miRNome (three 384-well plates) with 100

Sample-to-results workflow for miRNA biomarker discovery

normal samples and 100 breast cancer samples, which would

■■ tep 1: Prepare total RNA using the miRNeasy Serum/
S

take over 2 months of constant real-time PCR profiling, the method
outlined above can trim the miRNome to just 24 miRNAs of
interest, allowing 16 samples to be processed on one 384-well
Custom miScript miRNA PCR Array. This condensed panel
would, in contrast, take only take 1.5 days to assess, providing
substantial time and cost savings.

Plasma Kit.

■■ tep 2: Randomly choose 30 normal samples and 30
S
disease samples.

■■ Step 3: Combine the RNA to make 3 normal pools and 3
disease pools. Each pool contains RNA from 10 samples.

■■ tep 4: Perform reverse transcription (using the
S
Biomarker discovery workflow

miScript II RT Kit and miScript HiSpec Buffer), followed

Purify RNA from multiple samples

by real-time PCR quantification (using the miScript
SYBR® Green PCR Kit) on each of the pools. Just 5 µl
total RNA from the miRNeasy Serum/Plasma Kit provides
enough cDNA to profile the entire Human miRNome

Pool samples

3 x normal
sample pools
and 3 x
disease
sample pools

miScript miRNA PCR Array.

■■ Step 5: Analyze the data and choose the miRNA assays
for a Custom miScript miRNA PCR Array.

■■ tep 6: Profile all samples in the cohort using the Custom
S
Reverse transcription

miScript miRNA PCR Array.

Prepare PCR with miScript miRNA PCR Array

Conclusion
The study of circulating miRNAs in blood has the potential to
reveal insights into the function of miRNAs, their transport in the
or

cell, and their role in gene regulation in disease and nondisease
states. In addition, the use of circulating miRNAs as biomarkers
may be a useful, practical application of this knowledge. The
nature of circulating miRNAs results in several challenges in

Perform real-time PCR
1.E+001

experimental research not faced when examining miRNA from

Delta Rn

1.E-000

sources of abundant RNA, such as cells and tissues. Special

1.E-001
1.E-002
1.E-003

0

4

8

12

16

20 24

28 32

considerations must be taken with purification, quality control,

36 40

Cycle number

experimental controls, normalization, and experimental
design. In this paper, we have outlined experimentally verified

Analyze results, choose assays for
Custom miScript miRNA PCR Array

researcher to get projects up and running as rapidly as possible

10-1

10-5
10-4
10-3
10-2

and ensure reliable, easy-to-interpret data.

10-2
10-3
10-4
10-5

10-1

10-6
1

recommendations and QIAGEN kit solutions that enable the

1

Normal breast

p-Value for fold change

10-6

-4

-3

-2

-1

0

1

2

3

4

10-6

10-5

Fold change ratio [log2]

10-4

10-3

10-2

10-1

1

Breast tumor

Profile individual samples on

Scientific article

www.qiagen.com

9

References
1. Wu, H. et al. (2007) miRNA profiling of naïve, effector and memory CD8 T cells. PLoS One 2, e1020.
2. Yang, W. et al. (2005) Modulation of microRNA processing and expression through RNA editing by ADAR deaminases. Nat. Struct. Mol. Biol. 13, 13.
3. Mitchell, P.S. et al. (2008) Circulating microRNAs as stable blood-based markers for cancer detection. Proc. Natl. Acad. Sci. USA 105, 10513.
4. Chim, S.S. et al. (2008) Detection and characterization of placental microRNAs in maternal plasma. Clin. Chem. 54, 482.
5. Arroyo, J.D. et. al. (2011) Argonaute2 complexes carry a population of circulating microRNAs independent of vesicles in human plasma, Proc. Natl. Acad. Sci.
USA 108, 5003.
6. Vickers, K.C. et. al. (2011) MicroRNAs are transported in plasma and delivered to recipient cells by high-density lipoproteins. Nat. Cell Biol. 13, 423.
7. Wang, K. et al. (2010) Export of microRNAs and microRNA-protective protein by mammalian cells. Nucleic Acids Res. 38, 7248.
8. Hunter, M.P. (2008) Detection of microRNA expression in human peripheral blood microvesicles. PLoS One 3, e3694.
9. Mestdagh, P. et al. (2009) A novel and universal method for microRNA RT-qPCR data normalization. Genome Biology 10, R64.

Ordering Information
Product

Contents

Cat. no.

miRNeasy Serum/Plasma Kit (50)

Columns, plasticware, and reagents for 50 preps

217184

miRNeasy Serum/
Plasma Spike-In Control

10 pmol C. elegans miR-39 miRNA mimic spike-in control for serum/
plasma samples

219610

miRNeasy Micro Kit (50)


217084

miRNeasy Mini Kit (50)


217004

miRNeasy 96 Kit (4)

Columns, plasticware, and reagents for 4 x 96 preps

217061

RNeasy Protect Animal Blood Kit (50)


73224

PAXgene Blood miRNA Kit (50)


763134

miScript II RT Kit (12)

Reagents for 12 x 20 μl cDNA synthesis reactions

218160

miScript II RT Kit (50)

Reagents for 50 x 20 μl cDNA synthesis reactions

218161

miScript SYBR Green PCR Kit (200)

Reagents for 200 x 50 μl PCRs

218073

miScript SYBR Green PCR Kit (1000)

Reagents for 1000 x 50 μl PCRs

218075

miScript PCR Starter Kit (80)

Reagents for 10 x 20 μl cDNA synthesis reactions and 80 x 25 μl PCRs

218193

miScript Primer Assay (100)

miRNA-specific primer for 100 x 50 μl PCRs

Varies*

miRNome miScript miRNA PCR Array

miRNome panels of miRNA assays

331222

miScript miRNA PCR Array

Pathway or disease panels of miRNA assays

331221


Custom panels of miRNA assays

331231

* Visit GeneGlobe to search for and order these products (www.qiagen.com/GeneGlobe).

For up-to-date licensing information and product-specific disclaimers, see the respective QIAGEN kit handbook or user manual.
QIAGEN kit handbooks and user manuals are available at www.qiagen.com or can be requested from QIAGEN Technical Services
or your local distributor.

For more information on QIAGEN’s miRNA research portfolio, visit www.qiagen.com/Serum-Plasma.
Trademarks: QIAGEN®, QIAzol®, RNAprotect®, RNeasy®, GeneGlobe®, MinElute®, miScript®, QuantiTect® (QIAGEN Group); PAXgene® (PreAnalytiX GmbH); SYBR® (Molecular Probes, Inc.); Microsoft®, Excel® (Microsoft, Inc.).
1072963 06/2012 © 2012 QIAGEN, all rights reserved.

Australia n 1-800-243-800
Austria n 0800-281011
Belgium n 0800-79612
Brazil n 0800-557779
Canada n 800-572-9613
China n 800-988-0325
Denmark n 80-885945

Finland n 0800-914416
France n 01-60-920-930
Germany n 02103-29-12000
Hong Kong n 800 933 965
India n 1-800-102-4114
Ireland n 1800 555 049
Italy n 800-787980

Japan n 03-6890-7300
Korea (South) n 080-000-7145
Luxembourg n 8002 2076
Mexico n 01-800-7742-436
The Netherlands n 0800-0229592
Norway n 800-18859
Singapore n 1800-742-4368

Spain n 91-630-7050
Sweden n 020-790282
Switzerland n 055-254-22-11
Taiwan n 0080-665-1947
UK n 01293-422-911
USA n 800-426-8157
www.qiagen.com

Whitepaper serumplasma

More Related Content

What's hot (20)

Viewers also liked (7)

Similar to Whitepaper serumplasma (20)

More from Elsa von Licy (20)

Whitepaper serumplasma