Original articles

Improved protocol for plasma microRNA extraction and comparison of

commercial kits
Harshini Sriram1, Twinkle Khanka1, Shweta Kedia1, Priyanka Tyagi1, Sitaram Ghogale1, Nilesh Deshpande1, Gaurav Chatterjee1, Sweta
Rajpal1, Nikhil V Patkar1, Papagudi G Subramanian1, Sumeet Gujral1, Syed Hasan2, Prashant R Tembhare*1
1Hematopathology Laboratory, ACTREC, Tata Memorial Centre, Homi Bhabha National Institute, Kharghar, Navi Mumbai, India
2Hasan Laboratory, ACTREC, Tata Memorial Centre, Homi Bhabha National Institute, Kharghar, Navi Mumbai, India

*Corresponding author: [email protected]

Abstract
Introduction: MicroRNAs are small, non-coding RNA molecules that are becoming popular biomarkers in several diseases. However, their low abun-
dance in serum/plasma poses a challenge in exploiting their potential in clinics. Several commercial kits are available for rapid isolation of microRNA
from plasma. However, reports guiding the selection of appropriate kits to study downstream assays are scarce. Hence, we compared four commer-
cial kits to evaluate microRNA-extraction from plasma and provided a modified protocol that further improved the superior kit’s performance.
Materials and methods: We compared four kits (miRNeasy Serum/Plasma, miRNeasy Mini Kit from Qiagen; RNA-isolation, and Absolutely-RNA
MicroRNA Kit from Agilent technologies) for quality and quantity of microRNA isolated, extraction efficiency, and cost-effectiveness. Bioanalyzer-
based Agilent Small RNA kit was used to evaluate quality and quantity of microRNA. Extraction efficiency was evaluated by detection of four endo-
genous control microRNA using real-time-PCR. Further, we modified the manufacturer’s protocol for miRNeasy Serum/Plasma kit to improve yield.
Results: miRNeasy Serum/Plasma kit outperformed the other three kits in microRNA-quality (P < 0.005) and yielded maximum microRNA-quan-
tity. Recovery of endogenous control microRNA i.e. hsa-miR-24-3p, hsa-miR-191-5p, hsa-miR-423-5p and hsa-miR-484 was higher as well. Modifi-
cation with the inclusion of a double elution step enhanced yield of microRNA extracted with miRNeasy Serum/Plasma kit significantly (P < 0.001).
Conclusion: We demonstrated that miRNeasy Serum/Plasma kit outperforms other kits and can be reliably used with a limited plasma quantity. We
have provided a modified microRNA-extraction protocol with improved microRNA output for downstream analyses.
Keywords: microRNA; plasma; commercial reagent kits

Submitted: May 19, 2021 Accepted: August 14, 2021

Introduction
MicroRNA (miRNA) comprise a class of short (~22 tions, miRNA is protected from endogenous RNase
nucleotides), endogenous non-coding RNA that activity by its association with vesicles or proteins
are implicated in post-transcriptional regulation of such as Ago2 or other RNA-binding proteins (8-10).
protein expression (1). Several studies have high- Therefore, cfmiRNA offer an unbiased and viable
lighted the clinical utility of miRNA in the diagno- alternative to existing strategies to ascertain dis-
sis and prognostication of disease (2-4). In particu- ease condition.
lar, circulating or cell-free miRNA (cfmiRNA) have However, the reliable application of cfmiRNA in
gained importance as non-invasive biomarkers for clinics is still limited due to the lack of robust re-
screening and monitoring of both solid and hae- producibility in clinical samples. One of the major
matological malignancies. MicroRNA in biofluids challenges of the clinical applications of cfmiRNA
like serum/plasma exhibit disease specificity and studies is the inadequate understanding of many
has remarkable stability (5-7). In cell-free condi- preanalytical variables, including sample storage

https://doi.org/10.11613/BM.2021.030705 Biochem Med (Zagreb) 2021;31(3):030705

©Copyright by Croatian Society of Medical Biochemistry and Laboratory Medicine. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creative- 1
commons.org/licenses/by/4.0/) which permits users to read, download, copy, distribute, print, search, or link to the full texts of these articles in any medium or format and to remix, transform and build upon
the material, provided the original work is properly cited and any changes properly indicated​.
Sriram H. et al. Comparative analysis of plasma miRNA extraction kits

and RNA isolation from small volumes of clinical samples collected with EDTA anticoagulant. Whole
samples. To date, several kits are commercially blood samples were collected from 30 healthy do-
available to facilitate the rapid extraction of cfmiR- nors. Written informed consents were obtained
NA. However, studies evaluating the effectiveness from all volunteers. The experimental design im-
of commercially available kits with clinical speci- plemented in this study has been illustrated in Fig-
mens are scarce (11-13). Hence, the present study ure 1.
was aimed to assess the performance of four com- According to the protocol approved by the Institu-
monly available commercial kits for isolation of tional Ethics Committee (Tata Memorial Centre –
cfmiRNA from small volumes of human plasma Institutional Ethics Committee III), approximately 3
samples. We also modified the manufacturer’s mL of whole blood samples were collected in 5 mL
protocol for the kit with superior performance to K2EDTA vacutainer tubes (BD vacutainer, Becton
improve the miRNA yield further. Dickinson, Plymouth, UK). The samples were spun
at 1500xg for 15 min at 4 ºC to separate plasma.
Without disturbing the buffy coat, the clear super-
Materials and methods
natant was transferred to a fresh 2 mL cryotube
Subjects (Tarsons, Saltlake, Kolkata, India) and inspected for
haemolysis. The samples with no visual signs of
This was a prospective study focusing on the haemolysis were immediately frozen at - 80 ºC for
standardization of miRNA extraction from plasma 4-6 months.

Plasma separation at 1500xg for 15 min

Freeze at -80 °C

Thaw on ice before use

miRNa extraction

Optimization miRNeasy Absolutely

miRNeasy RNA RNA
Serum/
Mini Kit Isolation Kit MicroRNA
Plasma Kit
Kit
Manufacturer’s
Protocol (P1)

7 volumes of Small RNA quantification – Bioanalyzer

denaturing buffer (P2)

Additional wash with

buffer RWT (P3)

Double elution (P4) Check extraction efficiency – qPCR

Figure 1. Schematic view of the experimental design and workflow. miRNA - microRNA. qPCR - real-time polymerase chain reaction.

Biochem Med (Zagreb) 2021;31(3):030705 https://doi.org/10.11613/BM.2021.030705

2
Sriram H. et al. Comparative analysis of plasma miRNA extraction kits

Materials into the homogenate. Subsequently, 200 μL of

chloroform was added to the lysate and incubated
We assessed four commercially available micro/ for 2 min at room temperature. The solution was
small RNA extraction kits for their quality, yield, ex- centrifuged for 15 min at 12,000xg at 4 °C. The up-
traction efficiency, and cost-effectiveness. The per aqueous layer was transferred to a fresh mi-
study included miRNeasy Serum/Plasma kit (Qia- crofuge tube and mixed with 1.5 volumes of 100%
gen, Hilden, Germany), miRNeasy Mini kit (Qiagen, ethanol. The solution was transferred into an RNe-
Hilden, Germany), RNA isolation kit (Agilent Tech- asy MinElute spin column (Qiagen, Hilden, Germa-
nologies, Santa Clara, USA), and Absolutely RNA ny) and centrifuged at 8000xg for 15 seconds at
MicroRNA kit (Agilent Technologies, Santa Clara, room temperature. The spin column was washed
USA). RNeasy MinElute Cleanup Kit (Qiagen, twice, once each with the two wash buffers pro-
Hilden, Germany) was obtained for use with vided (RWT and RPE), and then with 80% ethanol.
miRNeasy Mini kit. For use as spike-in or exoge- Finally, miRNA was eluted in 14 μL RNase-free wa-
nous control 5’-phosphorylated synthetic RNA oli- ter.
go C. elegans miRNA cel-miR-39-3p (Sigma-Aldrich,
St. Louis, Missouri, USA) was obtained. Agilent Kit 2. miRNeasy Mini kit
Small RNA kit (Agilent Technologies, Santa Clara, Briefly, 700 μL QIAzol lysis Reagent was mixed with
USA) was procured for miRNA quantification. The 200 μL of plasma and incubated for 5 min at room
efficiency of each protocol was evaluated with temperature. A total of 5 pM of cel-miR-39-3p was
TaqMan technology-based real-time polymerase spiked into the homogenate. Afterwards, 140 μL of
chain reaction (qPCR) using TaqMan Advanced chloroform was added to the lysate and incubated
miRNA cDNA Synthesis kit, TaqMan Fast Advanced for 2 min at room temperature. The solution was
Master Mix, and TaqMan Advanced miRNA Assays centrifuged for 15 min at 12,000xg at 4 °C. The up-
(Applied Biosytems, Thermo Fisher Scientific, per aqueous phase was transferred to a fresh mi-
Waltham, USA). crofuge tube and mixed with equal volume of 70%
ethanol. The solution was transferred into an RNe-
Methods asy Mini column and centrifuged at 8000xg for 15
sec. The flow-through was collected into a fresh
Plasma preparation microfuge tube and mixed well with 0.65 volumes
of 100% ethanol. The sample was transferred into
Plasma samples were allowed to thaw completely
a fresh RNeasy MinElute spin column and centri-
on ice and centrifuged at 3000xg for 5 min at 4 °C
fuged at 8000xg for 15 sec. The spin-column was
to remove any cryoprecipitate. A total of 200 μL of
subjected to two washes, once with each of the
plasma was utilized per extraction.
buffers provided (RWT and RPE) and then with
miRNA extraction 80% ethanol. Subsequently, miRNA was eluted in
14 μL RNase-free water by centrifugation of the
RNA extraction was performed in accordance with
spin column at 8000xg.
the manufacturer’s protocol for each of the four
kits. For all miRNA isolations, 5 pM (5.44 x108 cop- Kit 3. RNA isolation kit
ies or 0.54 x108 copies/µL) of the synthetic cel-miR- The RNA isolation kit follows a column-free extrac-
39-3p was spiked into the sample before extrac- tion protocol based on alcohol precipitation of
tion. RNA. In short, 200 μL of plasma was incubated
Kit 1. miRNeasy Serum/Plasma kit with 2 mL of denaturing solution (solution D). A to-
tal of 5pM of cel-miR-39-3p was spiked into the
Briefly, 5 volumes of QIAzol lysis Reagent (Qiagen,
homogenate. This was followed by the addition of
Hilden, Germany) were mixed with 200 μL of plas-
200 μL of 2M sodium acetate (pH 4.0), 2 mL of phe-
ma and incubated at room temperature for 5 min.
nol (pH 5.3-5.7), and 400 μL of chloroform-isoamyl
Post incubation, 5 pM cel-miR-39-3p was spiked
alcohol to the lysate. The solution was mixed well

https://doi.org/10.11613/BM.2021.030705 Biochem Med (Zagreb) 2021;31(3):030705

3
Sriram H. et al. Comparative analysis of plasma miRNA extraction kits

and centrifuged for 15 min at 12,000xg at 5-10 °C. reagent being the maximum holding capacity of a
The upper aqueous phase was transferred to a 2 mL microfuge tube per 200 μL of plasma used.
fresh centrifuge tube. One volume of isopropanol Next, the column was subjected to an increased
was added to the aqueous phase, mixed well, and number (2x) of washes with the wash buffer (buff-
centrifuged for 30 min at 10,000xg. The pellet was er RWT) prior to the elution step (Protocol3/P3). Fi-
washed with 75% ethanol, air-dried, and resus- nally, the elution step of the protocol was modi-
pended in 50 μL RNase-free water. fied by incorporating a double elution protocol
Kit 4. Absolutely RNA MicroRNA kit (Protocol4/P4). A 10 min bench-top incubation
with RNase-free water was allowed before miRNA
Briefly, 200 μL of lysis buffer containing elution. The eluent was added onto the column
β-mercaptoethanol was added to 200 μL of plas- again and eluted a second time to ensure efficient
ma. The contents were mixed well to homogenize. elution of all membrane-bound RNA.
A total of 5pM of cel-miR-39-3p was spiked into
the homogenate, followed by the addition of one Small RNA quantification
volume of phenol-chloroform (1:1). The solution
The concentration of miRNA in the elute was
was mixed well and centrifuged for 4 min at maxi-
measured on a Bioanalyzer 2100 (Agilent Technol-
mum speed at room temperature. The upper
ogies, Santa Clara, USA) using the Agilent Small
aqueous phase was transferred to a fresh mi-
RNA kit. Instrument set-up, reagent preparation,
crofuge tube. One volume of chloroform-isoamyl
and small RNA assay were performed according to
alcohol (24:1) mixture was added and centrifuged
the manufacturer’s instructions. The small RNA re-
for 3 min at maximum speed. The resulting aque-
gion and miRNA region were arbitrarily defined as
ous phase was transferred into a prefilter spin cup
fragments between 10-150 nucleotides and 14-30
provided and centrifuged for 3 min at maximum
nucleotides, respectively.
speed. The filtrate was collected, mixed with 1.25
volumes of 100% ethanol, and transferred into an Given the difference in elution volumes across the
RNA-binding spin cup. The RNA-binding cup was kits being evaluated, data is represented as the to-
centrifuged for 1 min at full speed. The spin cup tal amount of miRNA. Total amount of miRNA per
was washed with Low-Salt buffer provided and extraction was calculated as follows: Total amount
subjected to on-column DNA digestion for 15 min of miRNA = concentration x volume in mL cDNA
at 37 °C. The spin cup was then washed with the synthesis.
High-Salt buffer provided. Furthermore, 50 μL of Reverse transcription (RT) was performed using
RNase-free water was added onto the column and TaqMan Advanced miRNA cDNA Synthesis Kit ac-
incubated on the bench-top for 1 min. Finally, cording to the manufacturer’s protocol. Briefly, 5
miRNA was eluted by centrifugation of the spin ng of the extracted RNA was reverse transcribed
column at maximum speed for 1 min. using universal RT primers supplied in the kit. The
cDNA was stored at - 20 ºC until further use.
Modifications to the miRNeasy Serum/Plasma kit
Modifications to the manufacturer’s protocol (Pro- Real-time PCR
tocol1/P1) were made to optimize miRNA extrac- MicroRNA quantitation was carried out using the
tion. Cell free miRNA was extracted from the plas- TaqMan Advanced miRNA Assays and TaqMan Fast
ma of 20 healthy donors collected and frozen pre- Advanced Master Mix on a QuantStudio real-time
viously. All extractions were performed in dupli- PCR system (Applied Biosystems, Thermo Fisher
cates to access reproducibility and intra-sample Scientific, Waltham, USA). Real time PCR reactions
variability. The protocol was modified at different were set up for the spike-in control and four en-
stages. First, the ratio of denaturing buffer to plas- dogenous miRNA controls - hsa-miR-24-3p, hsa-
ma volume was increased from the suggested 5:1 miR-191-5p, hsa-miR-423-5p and hsa-miR-484. In-
to 7:1 (Protocol2/P2), with 7 volumes of denaturing dividual qPCR assays were performed in triplicates

Biochem Med (Zagreb) 2021;31(3):030705 https://doi.org/10.11613/BM.2021.030705

4
Sriram H. et al. Comparative analysis of plasma miRNA extraction kits

with a total reaction volume of 10 μL. The assays Results

were performed according to the manufacturer’s
instructions. miRNA quality and quantity assessment

Statistical analysis Our data revealed that the quality of miRNA from
the miRNeasy Serum/Plasma kit was better than
Statistical analyses were performed using Graph- the other 3 kits (P < 0.005). The quality of miRNA
Pad Prism version 8.0 for Windows (GraphPad extracted using the RNA Isolation kit and Abso-
Software, San Diego, USA). The results were ex- lutely RNA MicroRNA kit from Agilent Technolo-
pressed as the median and interquartile range gies showed a significant decrease (P < 0.001) in
(IQR). The difference between quality (percentage comparison with the miRNeasy Serum/Plasma kit.
purity) and quantity of the miRNA yield and en- The miRNeasy Mini kit and Absolutely RNA Micro-
dogenous control recovery between the different RNA kit produced yields of comparable quality,
kits were evaluated using one way ANOVA. Tukey’s with no significant difference from one another (P
multiple comparison test was used post hoc for = 0.661). The results of the comparison of miRNA
multiple testing and P values were reported. quality isolated from the kits are given in Table 1.
A paired sample t-test was performed to compare Similarly, we observed that the miRNeasy Serum/
each protocol modification to that of the manu- Plasma kit yielded maximum miRNA upon elution.
facturer’s protocol. Intra-assay variability was de- There was no significant difference in miRNA
termined with the variation (CV) coefficient for the yields between this and the Absolutely RNA Micro-
manufacturer’s protocol and the three modifica- RNA kit (P = 0.198). We did not observe a signifi-
tions and expressed as a percentage (CV%). The cant difference in miRNA quantity between Qia-
level of significance for all tests performed was de- gen’s miRNeasy Mini kit and RNA Isolation kit from
fined as P < 0.05. Agilent Technologies (P = 0.237). However, the

Table 1. Qualitative and quantitative comparison of miRNA extraction kits

miRNeasy Serum / Absolutely

Variable miRNeasy Mini kit RNA Isolation kit P-value
Plasma kit RNAMicroRNA kit
< 0.001*
< 0.001†
miRNA quality 77.50 72.00 42.50 73.50 < 0.001‡
(% miRNA in small RNA) (74.00-83.00) (68.50-76.25) (34.75-50.25) (68.00-77.25) < 0.001§
0.661ǁ
< 0.001¶
< 0.001*
< 0.001†
miRNA yield < 0.198‡
19.20 17.30 16.10 18.20
(total amount of miRNA
(16.80-20.30) (14.10-18.10) (14.00-17.70) (17.40-19.20) < 0.237§
in ng)
< 0.001ǁ
< 0.001¶
Results are represented as median and interquartile range (Q1-Q3). The differences between the four kits were estimated using
one way ANOVA. Multiple comparisons were made using Tukey’s test. For all tests, P < 0.05 was considered significant. *Comparing
miRNeasy Serum/Plasma and miRNeasy Mini kit; †comparing miRNeasy Serum/Plasma and RNA Isolation kit; ‡comparing miRNeasy
Serum/Plasma and Absolutely RNA MicroRNA kit; §comparing miRNeasy Mini and RNA isolation kit; ǁcomparing miRNeasy Mini and
Absolutely RNA MicroRNA kit; ¶comparing Absolutely RNA MicroRNA kit and RNA Isolation.

https://doi.org/10.11613/BM.2021.030705 Biochem Med (Zagreb) 2021;31(3):030705

5
Sriram H. et al. Comparative analysis of plasma miRNA extraction kits

miRNeasy Serum/Plasma kit yielded higher quan- Isolation kit demonstrated the lowest recovery as
tities of miRNA (P < 0.001) in comparison to the reflected by higher Ct values for the endogenous
abovementioned kits (Table 1). Additionally, the miRNAs. The results of the comparison of endoge-
two extraction kits from Agilent Technologies also nous miRNA recovery between the kits are pre-
exhibited a significant difference in the total sented in Table 2.
amount of miRNA extracted (P < 0.001).
Optimization of miRNeasy Serum/Plasma kit
Detection of endogenous control miRNA
Initial results suggested that miRNeasy Serum/
All four kits recovered the spike-in miRNA and en- Plasma kit (Qiagen, Hilden, Germany) outper-
dogenous control miRNAs. However, the extent of formed the other isolation kits and provided supe-
recovery varied considerably. Detection of endog- rior quality elute with a higher quantity of miRNA.
enous miRNA is represented by their normalized The recovery of endogenous control miRNA was
quantitation cycle (Ct) value across the 4 kits (Ta- also better with the miRNeasy Serum/Plasma kit
ble 2). Our results indicated that the overall recov- than the other three tested. We hypothesized that
ery of the control miRNAs was greater with the a few modifications to the manufacturer’s proto-
miRNeasy Serum/Plasma kit. Additionally, the RNA col for miRNeasy Serum/Plasma kit could further

Table 2. Endogenous miRNA recovery comparison of miRNA extraction kits

miRNeasy Serum/ Absolutely RNA

Variable miRNeasy Mini kit RNA Isolation kit P-value
Plasma kit MicroRNA kit
< 0.001*
< 0.001†
Recovery of 18.26 20.46 21.46 19.11 0.013‡
hsa-miR-24-3p (Ct) (17.27-18.84) (19.41-21.85) (20.19-22.84) (18.30-19.85) 0.012§
0.001ǁ
< 0.001¶
0.008*
< 0.001†
Recovery of
19.33 20.27 21.58 20.18 0.483‡
hsa-miR-191-5p
(18.18-20.63) (19.13-21.61) (20.43-22.82) (18.46-20.63) 0.010§
(Ct)
0.269ǁ
< 0.001¶
0.142*
0.004†
Recovery of
18.72 19.81 20.47 19.97 0.521‡
hsa-miR-423-5p
(17.99-20.83) (18.79-21.50) 19.07-21.54 18.27-20.90 0.528§
(Ct)
0.861ǁ
0.146¶
< 0.001*
< 0.001†
Recovery of 18.35 20.15 21.10 19.12 0.110‡
hsa-miR-484 (Ct) 17.28-19.21 19.26-21.98 19.33-22.60 18.00-20.04 0.856§
0.001ǁ
< 0.001¶
Results are represented as median and interquartile range (Q1-Q3). The differences between the four kits were estimated using
one way ANOVA. Multiple comparisons were made using Tukey’s test. For all tests, P < 0.05 was considered significant. *Comparing
miRNeasy Serum/Plasma and miRNeasy Mini kit. †comparing miRNeasy Serum/Plasma and RNA Isolation kit. ‡comparing miRNeasy
Serum/Plasma and Absolutely RNA MicroRNA kit. §comparing miRNeasy Mini and RNA isolation kit. ǁcomparing miRNeasy Mini
and Absolutely RNA MicroRNA kit. ¶comparing Absolutely RNA MicroRNA kit and RNA Isolation. Ct – quantification cycle value
normalized using cel-miR-39-3p spike-in control.

Biochem Med (Zagreb) 2021;31(3):030705 https://doi.org/10.11613/BM.2021.030705

6
Sriram H. et al. Comparative analysis of plasma miRNA extraction kits

improve miRNA yield. Hence, we implemented a 0.001) while P4 improved the quality of the yield
few modifications as described in the “Materials significantly (P = 0.007).
and methods” section. A similar trend was noticed in the relative variabili-
The results of a comparison between the modi- ty between the quantity of miRNA extracted using
fied protocols (P2-P4) with the manufacturer’s original and modified protocols. However, we ob-
original protocol (P1) are given in Table 3. Intra- served a greater variability across all four proto-
sample variability was calculated for quality and cols. Average CV% for the quantity of miRNA iso-
quantity of miRNA extracted by the original and lated from P1 was 7.62%, P2 - 8.81%, P3 - 10.54%,
modified protocols and is expressed as average and P4 - 6.29%. In comparison we noticed no sig-
CV%. The quality of miRNA isolated from the nificant difference in the quantity of miRNA yield-
original protocol (P1) showed variability (CV%) of ed from P1 and P2 (P = 0.189). On the other hand,
6.92%, and that of modified protocol P2 was P3 yielded lower quantities of miRNA (P = 0.003)
5.67%, P3 - 7.06%, and P4 - 5.04%. There was no while P4 significantly improved the miRNA yield
significant difference between the quality of miR- upon extraction (P < 0.001). Thus, P4-modification
NA extracted from P1 and P2 (P = 0.061). In com- demonstrated the lowest variability in the quality
parison, P3 had a marked decrease in quality (P < and quantity of miRNA extracted.

Table 3. Assessment of miRNeasy Serum/Plasma protocol amendments

Protocol P1 P2 P3 P4 P-value
0.061*
miRNA quality 79.00 77.25 71.00 81.25
< 0.001†
(% miRNA in small RNA) (73.13-81.75) (72.00-81.38) (68.25-74.75) (76.13-83.00)
0.007‡
0.189*
miRNA yield 19.25 18.25 17.80 21.35
0.003†
(Total amount of miRNA in ng) (16.93-20.75) (17.05-19.68) (16.88-19.25) (18.40-22.58)
< 0.001‡
Protocol P1 is the manufacturer’s original protocol for miRNeasy Serum/Plasma kit. P2–P4 are modified protocols tested to optimize
miRNA extraction from the kit. Results are represented as median and interquartile range (Q1-Q3). The differences between the four
kits were estimated using paired sample t-test. For all tests, P < 0.05 was considered significant. *Comparing P1 and P2. †comparing
P1 and P3. ‡comparing P1 and P4.

Discussion
In this study, we assessed four commercially avail- all parameters evaluated. This could be partly at-
able RNA isolation kits for their efficiency in ex- tributed to the duration of extraction or even re-
tracting good quality and quantity of miRNA. sidual salt contamination. However pure the rea-
The sensitivity of downstream assays of miRNA, in- gents may be, prolonged RNA isolation protocols
cluding qPCR and microarray, is largely affected by elevate the risk of RNase contamination, leading to
residual salts from denaturing and wash buffers degradation of miRNA. Additionally, this kit does
used in the extraction process. Due to the low not employ a filtering agent, such as silica gel or
abundance of miRNA in plasma, selecting an ideal beads, to ensure adequate removal of residual
miRNA isolation kit becomes crucial. In this study, chaotropic salts from buffers used. It relies on the
the RNA Isolation kit from Agilent Technologies principles of organic extraction and thus, may not
demonstrated relatively lower performance across be a very compelling choice for small RNA studies.

https://doi.org/10.11613/BM.2021.030705 Biochem Med (Zagreb) 2021;31(3):030705

7
Sriram H. et al. Comparative analysis of plasma miRNA extraction kits

On the other hand, our findings suggested that miRNA extracted. This increase in concentration
the miRNeasy Serum/Plasma kit provides relatively also reflected positively on the quality of miRNA
higher quality and quantity of miRNA when com- extracted.
pared to other kits. The ease of use and short du- Previous studies suggested that DNA contamina-
ration for extraction provide further advantages. tion in extracted miRNA could interfere with miR-
The quality of miRNA is a crucial factor as it im- NA detection (15). Hence, we further evaluated the
pacts downstream analyses, especially quantifica- effect of possible DNA carryover on the detection
tion. The RNA profile of plasma is depleted of tra- of endogenous control miRNA extracted using
ditional markers of RNA quality and integrity, in- modified (P4) protocol for miRNeasy Serum/Plas-
cluding large RNA species such as messenger RNA ma kit by TaqMan technology-based qPCR reac-
and the ribosomal RNA subunits. The RNA pool tions (data not shown). DNA carryover was quanti-
from peripheral blood is dominated by miRNA, fied using Qubit Double-Stranded DNA High Sen-
fragmented mRNA, small nucleolar RNA, and oth- sitivity Assay kit (Thermo Fisher Scientific,
er small RNA species (14). Hence, the 28s to 18s Waltham, USA). We noticed DNA carryover in the
rRNA ratio is not reliable in the assessment of range of 0.06-2.15 ng/µL. To evaluate DNA interfer-
cfmiRNA quality. In our study, this limitation was ence in the quantification of miRNA we compared
overcome by using the proportion of miRNA in to- the detection of endogenous control miRNAs with
tal small RNA as an indicator of cfmiRNA quality. and without reverse transcription. We observed
For this, the size range of small RNA and miRNA that signal was produced against all 4 miRNAs only
were arbitrarily defined as fragments between 10- upon reverse transcription. Endogenous miRNAs
150 and 14-30 nucleotides in length respectively. A were not detected in the RT control (subject to all
higher proportion of miRNA in small RNA indicat- steps of cDNA synthesis but for the addition of Re-
ed a greater quality of RNA isolated. verse Transcriptase), suggesting that a DNase I di-
One of the challenges of plasma miRNA studies is gestion is not essential under these circumstances.
that despite being one of the dominant RNA spe- Overall, the results of our study were in accord-
cies in liquid biopsies, miRNA yields from plasma ance with findings reported by Inés Moret et al.,
are much lower than solid tissues. This limitation Kathryn Wright et al., and Ari Meerson et al. (16-18).
demands kits that can maximize miRNA yield from These studies also compared various miRNA isola-
minimal input volumes of serum or plasma. tion kits and demonstrated that the miRNeasy Se-
Most recommended protocols provided with the rum/Plasma kit from Qiagen produces the most
commercial kits are generic and require additional favorable results. However, with the exception of
optimization to be reliably used. Optimization of the miRNeasy Serum/Plasma Kit, no other kit was
protocols is required to obtain high-quality, high- common between the current study and these
yielding miRNA from small volumes of input sam- studies.
ples in reproducible manner. The miRNeasy Se- We believe that such a study, among others, is es-
rum/Plasma kit showed superior performance sential in initiating miRNA-based diagnostics and
across parameters tested compared to the other prognostication. The novelty of this study lies in
kits evaluated in this study. Hence, we hypothe- the modification to optimize the manufacturer’s
sized that additional optimization of the protocol protocol. Our optimized protocol enhanced the
provided with miRNeasy Serum/Plasma kit could miRNA yield of the commercially available miRNe-
further improve the resulting miRNA yield. We as- asy Serum/Plasma kit. Combined, it could pave the
sessed an improvisation to the quick-start proto- way for validating miRNA as a cancer biomarker
col from the kit for plasma miRNA extraction. We and promoting its use in clinics. We acknowledge
observed that incorporating a double-elution step that the small sample size limits our study. Studies
with an on-column incubation of 10 minutes leads with a larger sample cohort can add value to the
to a significant increase in the concentration of validation of our results. Our study was focused on

Biochem Med (Zagreb) 2021;31(3):030705 https://doi.org/10.11613/BM.2021.030705

8
Sriram H. et al. Comparative analysis of plasma miRNA extraction kits

miRNA extraction from plasma samples, and simi- used to extract cfmiRNA from small volumes of
lar studies across other body fluids are necessary plasma. We have provided an optimized protocol
for the widespread use of miRNA in research and to improve miRNA isolation further for down-
diagnostics. stream analyses of mature miRNA.
In conclusion, we have demonstrated that miRNe- Potential conflict of interest
asy Serum/Plasma kit from Qiagen can be reliably
None declared.

References
1. O’Brien J, Hayder H, Zayed Y, Peng C. Overview of MicroR- 10. Wang K, Zhang S, Weber J, Baxter D, Galas DJ. Export of
NA Biogenesis, Mechanisms of Actions, and Circulati- microRNAs and microRNA-protective protein by mamma-
on. Front Endocrinol (Lausanne). 2018;9:402. https://doi. lian cells. Nucleic Acids Res. 2010;38:7248-59. https://doi.
org/10.3389/fendo.2018.00402 org/10.1093/nar/gkq601
2. Ell B, Mercatali L, Ibrahim T, Campbell N, Schwarzenbach H, 11. Bergallo M, Gambarino S, Martino S, Montin D, Montanari P,
Pantel K, et al. Tumor-induced osteoclast miRNA changes Galliano I, et al. Comparison of Two Available RNA Extracti-
as regulators and biomarkers of osteolytic bone metasta- on Protocols for microRNA Amplification in Serum Samples.
sis. Cancer Cell. 2013;24:542-56. https://doi.org/10.1016/j. J Clin Lab Anal. 2016;30:277-83. https://doi.org/10.1002/
ccr.2013.09.008 jcla.21848
3. Gao YC, Wu J. MicroRNA-200c and microRNA-141 as poten- 12. Li Y, Kowdley KV. Method for microRNA isolation from clini-
tial diagnostic and prognostic biomarkers for ovarian can- cal serum samples. Anal Biochem. 2012;431:69-75. https://
cer. Tumour Biol. 2015;36:4843-50. https://doi.org/10.1007/ doi.org/10.1016/j.ab.2012.09.007
s13277-015-3138-3 13. McAlexander MA, Phillips MJ, Witwer KW. Comparison of
4. Ren A, Dong Y, Tsoi H, Yu J. Detection of miRNA as non- Methods for miRNA Extraction from Plasma and Quantita-
invasive biomarkers of colorectal cancer. Int J Mol Sci. tive Recovery of RNA from Cerebrospinal Fluid. Front Genet.
2015;16:2810-23. https://doi.org/10.3390/ijms16022810 2013;4:83. https://doi.org/10.3389/fgene.2013.00083
5. Chen X, Ba Y, Ma L, Cai X, Yin Y, Wang K, et al. Characterizati- 14. Savelyeva AV, Kuligina EV, Bariakin DN, Kozlov VV, Ryabchi-
on of microRNAs in serum: a novel class of biomarkers for di- kova EI, Richter VA, et al. Variety of RNAs in Peripheral Blo-
agnosis of cancer and other diseases. Cell Res. 2008;18:997- od Cells, Plasma, and Plasma Fractions. Biomed Res Int.
1006. https://doi.org/10.1038/cr.2008.282 2017;2017:7404912. https://doi.org/10.1155/2017/7404912
6. Gilad S, Meiri E, Yogev Y, Benjamin S, Lebanony D, Yerushal- 15. Schamberger A, Orban TI. 3’ IsomiR species and DNA con-
mi N, et al. Serum microRNAs are promising novel biomar- tamination influence reliable quantification of microRNAs
kers. PLoS One. 2008;3:e3148. https://doi.org/10.1371/jour- by stem-loop quantitative PCR. PLoS One. 2014;9:e106315.
nal.pone.0003148 https://doi.org/10.1371/journal.pone.0106315
7. Mitchell PS, Parkin RK, Kroh EM, Fritz BR, Wyman SK, Pogo- 16. Moret I, Sanchez-Izquierdo D, Iborra M, Tortosa L, Navarro-
sova-Agadjanyan EL, et al. Circulating microRNAs as sta- Puche A, Nos P, et al. Assessing an improved protocol for
ble blood-based markers for cancer detection. Proc Natl plasma microRNA extraction. PLoS One. 2013;8:e82753.
Acad Sci U S A. 2008;105:10513-8. https://doi.org/10.1073/ https://doi.org/10.1371/journal.pone.0082753
pnas.0804549105 17. Wright K, de Silva K, Purdie AC, Plain KM. Comparison of
8. Köberle V, Pleli T, Schmithals C, Augusto Alonso E, Hau- methods for miRNA isolation and quantification from ovi-
penthal J, Bonig H, et al. Differential stability of cell-free cir- ne plasma. Sci Rep. 2020;10:825. https://doi.org/10.1038/
culating microRNAs: implications for their utilization as bio- s41598-020-57659-7
markers. PLoS One. 2013;8:e75184. https://doi.org/10.1371/ 18. Meerson A, Ploug T. Assessment of six commercial plasma
journal.pone.0075184 small RNA isolation kits using qRT-PCR and electrophore-
9. Li L, Zhu D, Huang L, Zhang J, Bian Z, Chen X, et al. Argona- tic separation: higher recovery of microRNA following ul-
ute 2 complexes selectively protect the circulating microR- tracentrifugation. Biol Methods Protoc. 2016;1:bpw003.
NAs in cell-secreted microvesicles. PLoS One. 2012;7:e46957. https://doi.org/10.1093/biomethods/bpw003
https://doi.org/10.1371/journal.pone.0046957

https://doi.org/10.11613/BM.2021.030705 Biochem Med (Zagreb) 2021;31(3):030705

9