Journal of Biological Methods | 2017 | Vol.

4(2) | e72
DOI: 10.14440/jbm.2017.184 Article

A cost-effective RNA extraction technique from animal

cells and tissue using silica columns
Mario D. Escobar*, Jason L. Hunt
Biology Department, College of Agriculture and Life Sciences, Brigham Young University Idaho, Rexburg, ID 83460, USA

*Corresponding author: Mario D. Escobar, Email: esc14004@byui.edu

Abbreviations used: βM, β-mercaptoethanol; G, times gravity; GITC, guanidine thiocyanate; GMMe, mink uterus endometrium epithelial cells; PBS, phosphate-buffered saline;
RIN, RNA integrity number; RQN, RNA quality number; RT-PCR, reverse transcriptase polymerase chain reaction

Received February 6, 2017; Revision received April 5, 2017; Accepted April 17, 2017; Published May 10, 2017

ABSTRACT

Ribonucleic acid (RNA) is widely used in molecular biology assays, and some of the most common assays include:
northern blotting and RT-PCR gene expression analysis. RNA is generally extracted by two methods: phenol-chloroform
or commercially available silica spin column kits. Phenol-chloroform extraction is generally more economical; however,
it produces hazardous byproducts, and leftover chemicals in the sample that can inhibit downstream applications. Com-
mercial kits usually have simple set ups and short preparation time; however, they can introduce a significant expense
to laboratory budgets. Here we have created a method to extract RNA using generic silica columns and readily available
reagents while maintaining a high yield and purity.
Keywords: cell culture, extraction, RNA, silica, tissue

INTRODUCTION it to low-cost silica columns while maintaining a high yield and quality
at an approximate cost of $0.48/sample.
Ribonucleic acid (RNA) is a ribose based nucleic acid that uses gua-
nine, uracil, adenine, and cytosine as nitrogenous bases. RNA is found
throughout the cell exhibiting functions that range from a template for MATERIALS AND METHODS
protein synthesis [1] to catalyzation of biological reactions [2]. With the
increased availability of cost effective and rapid genomic sequencing
resources, RNA has proved extremely valuable in the analysis of entire Cell culture
transcriptomes. The isolation and purification of RNA can be compli- GMMe cells obtained from the American Type Culture Collection
cated because of the presence of ribonuclease enzymes that rapidly (ATCC, Manassas, VA, CRL-267) were grown in Dulbecco’s Modified
degrade RNA. Although there are numerous ways to extract and isolate Eagle Medium/Nutrient Mixture F-12 (DMEM/F12; Caisson Labs,
RNA, most labs gravitate toward using phenol-chloroform extractions Smithfield, UT, DFL14-500ML) with 10% FBS (Fetal Bovine Serum;
or commercially available kits. Phenol-chloroform extractions use the Caisson Labs, Smithfield, UT, FBL02-500ML) at 37°C and 5% CO2.
property of phenol to separate the proteins from the nucleic acids, al- After 48 h of growth, the cells were trypsinized, counted and prepared
lowing for easy capture of the nucleic acids with ethanol precipitation. for RNA extraction.
Although, this method has become a staple for RNA extraction for
many scientists because of its high RNA yield and purity [3], the long Buffer creation
hands-on time, the hazardous waste, and sensitive steps make it a very Buffer composition is summarized in Table 1. Buffer A was com-
inefficient technique. Commercially available kits cut down on time and posed of 4 M guanidine thiocyanate (GITC; Promega, Madison, WI,
hazardous waste, and make use of very simple protocols which reduce V2791), 10 mM 2-(N-morpholino)ethanesulfonic acid (MES; Fisher
the likelihood of mistakes. However, these kits can be very expensive Scientific, Hampton, NH, BP300-100) pH 5.5, and 1% β-mercaptoethanol
with higher end kits costing upwards of $6.00 per extraction. This high (β-M; Sigma Aldrich, St. Louis, MO, M6250-10ML). β-M was added
cost can become a significant expense in the small lab’s budget. In our immediately before starting the extraction. Buffer B was composed of
article, we present a method using guanidine thiocyanate, a chaotropic 1 M GITC, 10 mM tris(hydroxymethyl)aminomethane (Tris; Fisher
salt with known denaturing abilities [4], to extract RNA and then bind Scientific, Hampton, NH, BP152-1) pH 7. Buffer C was composed of

How to cite this article: Escobar MD, Hunt JL. A cost-effective RNA extraction technique from animal cells and tissue using silica columns.
J Biol Methods 2017;4(2):e72. DOI: 10.14440/jbm.2017.184

www.jbmethods.org 1
POL Scientific
 Article
80% Ethanol (Decon Labs, King of Prussia, PA, V1005HC), and 10 mM 1 µl of 50 mM ethylenediaminetetraacetic acid (EDTA) was added.
Tris buffer pH 7. Following the addition of EDTA the reaction was incubated at 65°C
for 10 min and then stored at −80°C.
Table 1. Summary of buffer composition.
Tissue RNA extraction
Buffer A 4M Guanidine thiocyanate Rat (Rattus norvegicus) liver tissue was donated by Clair Eckersell
10 mM MES pH 5.5
(Brigham Young University - Idaho). Identical to the cell RNA extraction
protocols, both kit procedures followed the manufacturer’s instructions using
1% β-mercaptoethanol
30 mg of tissue with no modification and the subsequent RNA obtained
Buffer B 1M Guanidine thiocyanate
was stored at −80°C. For the silica column procedure, approximately 30
10 mM Tris pH 7 mg of tissue was placed in 2 ml centrifuge tubes (VWR, Radnor, PA,
Buffer C 80% Ethanol 89000-010) followed by the addition of 600 µl of Buffer A. Tissue was
10 mM Tris pH 7 disrupted using a homogenizer (Omni International, Kennesaw, GA,
TH115). The resultant mixture was centrifuged at 8000 G for 3 min and
the supernatant was transferred to a new 2 ml centrifuge tube and one
Cell RNA extraction volume of 70% ethanol was added. The mixture was then added to a
Cell culture media was carefully removed from flasks (T-160) and the silica column and centrifuged at 8000 G for 30 s. The resultant flow
cells were washed in 15 ml phosphate-buffered saline (PBS). Cells were through was discarded. The column was then washed with 600 µl of
then incubated for 10 min with 4 ml of trypsin (Caisson Labs, Smithfield, Buffer B followed by centrifugation at 8000 G for 30 s and the flow
UT, TRL02-100ML) and 12 ml of PBS at 37°C, and 5% CO2. After incu- through was discarded. The column was then washed twice with 500 µl
bation, the cell mixture was transferred into a new 50 ml centrifuge tube of Buffer C followed each time by centrifugation at 8000 G for 30 s and
(VWR, Radnor, PA, 89039-664) and centrifuged at 1600 G for 6 min at flow through was discarded. To clear any remaining buffer, the column
25°C. The supernatant was carefully removed and the subsequent cell was centrifuged for 2 min and then placed in a new 1.5 ml centrifuge
pellet was re-suspended with 3 ml of DMEM/F12 with 10% FBS. A tube. The RNA was captured by adding 30 µl of RNase free Mili-Q
10 µl aliquot sample of cell mixture was then retrieved and mixed with water to the column followed by centrifugation at 8000 G for 30 s. The
10 µl of trypan blue (Thermo Fisher, Waltham, MA, 15250-061). The solution containing the RNA was treated with DNase and then stored
cell-trypan blue mixture was then loaded into a Countess II FL Auto- at −80°C. Additionally, to test the efficacy of the silica column against
mated Cell Counter (Thermo Fisher, Waltham, MA, AMQAF1000) and varying amounts of tissue, we performed incremental extractions from
a cell count was calculated. The cell mixture was then aliquoted into starting tissue samples that ranged from 10 mg to 60 mg (Fig. 1).
new 1.5 ml centrifuge tubes containing approximately 4.5 × 105 cells
per tube. Tubes containing cells were then divided into three extraction
procedures that consisted of the proposed silica column procedure, an
RNeasy mini kit (Qiagen, Hilden, Germany, 74104) and a GenCatch total
RNA miniprep kit (Epoch Life Science, Missouri City, TX, 1660050).
Both kit procedures followed the manufacturer’s instructions with
no modification and the subsequent RNA obtained was stored at −80°C.
For the silica column procedure, cell culture media was carefully re-
moved and the remaining cells were washed in 1 ml PBS. Following
the washing and removal of PBS, 300 µl of Buffer A was added to
each well with cells. Cells were then incubated with Buffer A for 15
min. After incubation, 300 µl of 70% ethanol was added to each well
and the mixture was added to a silica column (Epoch Life Science,
Missouri City, TX, 1920-050) and centrifuged (Eppendorf, Hamburg,
Germany, 5424) at 8000 G for 30 s. The resultant flow through was
then discarded. The column was then washed again by the addition of
600 µl of Buffer B followed by centrifugation at 8000 G for 30 s. The
resultant flow through was again discarded. The column was then washed Figure 1. Mean RNA concentration from different initial amounts of
twice with 500 µl of Buffer C followed each time by centrifugation tissue. N = 9 samples per extraction procedure. Means with different
at 8000 G for 30 s and the flow through was discarded. Following the letters (A, B and C) are significant (P < 0.05).
washing steps, the column was centrifuged again for 2 min and then
placed in a new 1.5 ml centrifuge tube (VWR, Radnor, PA, 89000-
010). The RNA was captured by adding 30 µl of RNase free Mili-Q Fragment analysis and quantification
water to the column followed by centrifugation at 8000 G for 30 s. The RNA was sent to Idaho State University’s Molecular Research Core
solution containing the RNA was treated with DNase (Thermo Fisher, Facility for the quantification and the fragment analysis. RNA was quan-
Waltham, MA, EN0525) by adding 1 µl of 10 × reaction buffer and tified using an RNA fluorescence assay (Thermo Fisher, Waltham, MA,
1 µl of DNase. The reaction volume was then adjusted to 10 µl with Q32852) following the manufacturer’s instructions, with a fluorometer
RNase-free water and then incubated at 37°C for 30 min after which (Thermo Fisher, Waltham, MA, Qubit® 2.0), and using spectrometer

2 J Biol Methods | 2017 | Vol. 4(2) | e72

POL Scientific
 Article
(Thermo Fisher, Waltham, MA, Nanodrop 1000). RNA fragment was tegrity with a unit called the RNA quality number (RQN). Although
analyzed using a fragment analyzer (Advanced Analytics, Ankeny, IA, spectroscopy can be used to determine the concentration and purity of
Fragment Analyzer) and the results were analyzed by fragment analysis RNA it lacks the power to determine the integrity of the RNA. The RQN
software (Advanced Analytics, Ankeny, IA, PROSize). can estimate the integrity of the RNA using a proprietary algorithm
that uses the area before the 18 s peak, the total area of the 18 s and
28 s peaks, and the ratio of the 28 s and 18 s peaks as obtained from
RESULTS AND DISCUSSION the fragment analyzer graph. The RQN has been shown to correlate to
the RNA integrity number (RIN) [5]. The RIN numbers range from 0
Following RNA fragment analysis (Advanced Analytics, Ankeny, to 10. The number 0 is the lowest possible value and 10 is the highest
IA, PROSize), we found that our high efficiency and low cost extraction possible value. For RNA sequencing, where RNA integrity is essential for
technique for RNA from both cells and tissue showed little to no signs significant results, a RIN lower than 6.4 is detrimental to the results [6].
of degradation (Fig. 2). The RNA fragment analyzer scores RNA in-

Table 2. Extracted RNA concentration (ng/µl) and quality assessment (RQN).

Sample ID Concentration (ng/µl) 260/280 260/230 RQN

Tissue 1 1118.3 2.14 1.36 7.4
Tissue 2 513.14 2.13 0.99 7.3
Tissue 3 746.27 2.11 1.50 8.50
Cells 1 69.52 2.08 1.4 10.00
Cells 2 56.36 2.1 1.29 10.00
Cells 3 55.16 2.18 1.12 10.00

Figure 2. RNA fragment analysis. A-C. Tissues samples 1–3. D-F. Three independent cell cultures.

Although the cell samples showed a significantly smaller concentra- of variability of RNA quality and concentration when comparing enzy-
tion of RNA in comparison to the tissue (Table 2) we hypothesize that matically-rich liver tissue to high lipid adipose tissue in commercially
this difference in concentrations was the result of the variation in the available extractions.
number of cells obtained from cell culture vs the amount of cells used All 260/280 ratios for both cells and tissue were around 2.0 indicating
for the procedure or variability in the tissue samples. Our average cell that the composition of the eluent was RNA and not DNA or protein
count was around 4 × 105 cells, consistent with the average 12 well carryover. In contrast, the 260/230 ratios in tissue were below 1, indicating
plate which has around 4 × 105 cells [7], whereas 30 mg of rat liver has a contaminant, possibly a small amount of guanidine thiocyanate salt.
around 4.17 × 106 cells [8]. Additionally, we have observed spectrums However, the cells showed 260/230 ratios above one indicating better

J Biol Methods | 2017 | Vol. 4(2) | e72  3

POL Scientific
 Article
purity. We speculate that it would be possible to improve RNA yields and commercially available columns [11], yet the use of phenol and
by adding chelating agents (like EDTA), anti-foam agents (like isoamyl other hazardous reagents in these protocols can become a problem if
alcohol) or transcription inhibitors (like sodium lauroyl sarcosinate). not handled properly and could result in injury if the user is still new to
However, in order to keep the method simple and cost effective, these the protocol. Although guanidine thiocyanate is minimally hazardous,
reagents were omitted for a later study. it is considerably less hazardous than the aforementioned phenol and
The RNA results comparing the commercial kits to the silica column chloroform.
procedure showed no significant difference in the amount or quality of The procedural time of the protocol with an experienced user is
RNA (Fig. 3 and 4). These results support the claim that the proposed around 20 min, which is comparable to commercially available kits,
protocol can extract RNA at the same quality and level as the more whereas the preparation of the buffers, which can be done before hand,
expensive commercial kits. Although the silica column procedure was extended the time by another 15 min. Although a DNase step was
effective in extracting pure and consistent RNA, our data suggests that added to the procedure to better improve the quality of the RNA, we
there is a binding limit. After approximately 30 mg of tissue, additional found that omission of this step only resulted in very small amounts of
material did not significantly affect the amount of RNA extracted (Fig. 1). genomic DNA (unpublished data). With the same omission, we also
found small amounts of genomic DNA in commercial kits (unpublished
data). Overall, this article aims to aid new laboratories or scientists under
tight budget constraints to obtain the same quality of results for RNA
isolation as commercially available kits, while using less expensive yet
widely available reagents.

Figure 3. Mean concentration and quality of RNA from different cell

extraction procedures. N = 12 samples per extraction procedure.

Although the idea of binding RNA to free-floating silica particles

has been used in the past [9], the complexity and tediousness of using Figure 4. Mean concentration and quality of RNA from different tissue
unbounded silica particles, in contrast to using silica columns, can extraction procedures. N = 12 samples per extraction procedure.
overwhelm the average scientist foreign to the technique. Another pos-
sibility is creating in-house silica columns [10]. However, this is such
a vital step in the extraction procedure that without proper calibration, Acknowledgments
standardization, and sterility, large variations in yield and purity are
likely to compromise the results. There have been previous attempts We would like to thank Seth Ririe and Todd Kelson for their assistance
to create silica-based extraction protocols that rely on in-house buffers in manuscript editing. We would also like to thank Steve Christenson

4 J Biol Methods | 2017 | Vol. 4(2) | e72

POL Scientific
 Article
GEN 33: 17. doi: 10.1089/gen.33.2.09.
and Shane Ruebush for valuable discussion insights. This publication 6. Gallego Romero I, Pai AA, Tung J, Gilad Y (2014) RNA-seq: impact of
was made possible by an Institutional Development Award (IDeA) RNA degradation on transcript quantification. BMC Biol 12: 42-1186. doi:
from the National Institute of General Medical Sciences of the National 10.1186/1741-7007-12-42. PMID: 24885439
Institutes of Health under Grant #P20GM103408. 7. Useful numbers for cell culture [Internet]. Waltham, MA: ThermoFisher
Scientific. August 22, 2015 [Cited on January 11, 2017]. Available from: https://
www.thermofisher.com/us/en/home/references/gibco-cell-culture-basics/cell-
culture-protocols/cell-culture-useful-numbers.html.
References 8. Sohlenius-Sternbeck A (2006) Determination of the hepatocellularity number
1. Alberts B, Johnson A, Lewis J, Raff M, Roberts K, et al. (2002) From RNA to for human, dog, rabbit, rat and mouse livers from protein concentration
protein. In: Alberts B, Johnson A, Lewis J, Raff M, Roberts K, et al. Molecular measurements. Toxicol In Vitro 20: 1582-1586. doi: 10.1016/j.tiv.2006.06.003.
Biology of the Cell. New York: Garland Science. pp. 229-374. PMID: 16930941
9. Boom R, Sol CJ, Salimans MM, Jansen CL, Wertheim-van Dillen PM, et
2. Serganov A, Nudler E (2013) A decade of riboswitches. Cell 152: 17-24. doi:
al. (1990) Rapid and simple method for purification of nucleic acids. J Clin
10.1016/j.cell.2012.12.024. PMID: 23332744
Microbiol 28: 495-503. PMID: 1691208
3. Xiang X, Qiu D, Hegele RD, Tan WC (2001) Comparison of different methods 10. Borodina TA, Lehrach H, Soldatov AV (2003) DNA purification on homemade
of total RNA extraction for viral detection in sputum. J Virol Methods 94: 129- silica spin-columns. Anal Biochem 321: 135-137. PMID: 12963065
135. doi: 10.1016/S0166-0934(01)00284-1. PMID: 11337047
11. Gonzalez-Perez I, Armas Cayarga A, García de la Rosa I, Josefina González
4. Gordon JA (1972) Denaturation of globular proteins. Interaction of guanidinium González Y (2006) Homemade viral RNA isolation protocol using silica
salts with three proteins. Biochemistry 11: 1862-1870. PMID: 5063539 columns: a comparison of four protocols. Anal Biochem 360: 148-150. doi:
5. Wong KS, Pang HM (2013) Simplifying HT RNA quality & quantity analysis. 10.1016/j.ab.2006.10.022. PMID: 17113029

J Biol Methods | 2017 | Vol. 4(2) | e72  5

POL Scientific