400 – Tattersall et al.
Technical Brief
Comparison of Methods for Isolating High-Quality
RNA from Leaves of Grapevine
Elizabeth A.R. Tattersall,1 Ali Ergul,2 Fadi AlKayal,3 Laurent DeLuc,1
John C. Cushman,1 and Grant R. Cramer1*
Abstract: High concentrations of polyphenols and polysaccharides make it challenging to extract high-quality RNA
from grape organs. To determine an optimal protocol for grape leaves, 15 different methods of RNA extraction
were evaluated based on cost, time to complete the extraction, and quality of the RNA isolated. The addition of
specific compounds to the extraction buffer to remove polyphenols and polysaccharides is often critical for downstream applications such as polymerase chain reaction (PCR) and microarray hybridization. RNA quality was assessed
using spectrophotometric methods, formaldehyde-agarose gel electrophoresis, reverse transcription (RT)-PCR reactions, and Agilent 2100 Bioanalyzer. Large differences in RNA yield and quality among protocols were found.
Some protocols that are commonly used for other species did not yield usable RNA from grapevine leaves. The
optimum methods were Tris-lithium chloride, which, while relatively time-consuming, gave consistently high yields
of quality RNA at very low cost and was suitable for PCR and microarray hybridization, and RNeasy Midi + polyethylene glycol, which rapidly provided high-quality RNA, but with lower yields.
Key words: RNA isolation, leaf, grape, Vitis vinifera
Extraction of high-quality total RNA is essential for the
successful application of many molecular techniques, such
as reverse transcription polymerase chain reaction (RTPCR), cDNA library construction, and gene expression
profiling studies using microarrays. RNA is degraded rapidly by ribonucleases (RNases) and, therefore, must be
extracted quickly and efficiently (Sambrook et al. 1989).
Denaturing reagents such as phenol (Davies and Robinson
1996) or guanidine thiocyanate (John 1992, Salzman et al.
1999) and inhibitors of RNases such as aurintricarboxylic
acid (ATA) (Franke et al. 1995, Lewinsohn et al. 1994) are
often added to extraction buffers.
Extraction of high-quality RNA from the leaves of
woody plants, such as grapevine, is particularly challenging because of high concentrations of polysaccharides,
polyphenols, and other secondary metabolites (Loulakakis
et al. 1996, Salzman et al. 1999). When applied to grape
leaves, some RNA extraction methods yield pellets that
are poorly soluble, indicating the presence of unknown
contaminants, whereas others are gelatinous, indicating
the presence of polysaccharides (Tesniere and Vayda
1991). Other procedures yield reddish pellets indicative of
phenolic contamination (Bahloul and Burkard 1993, Katterman and Shattuck 1983). RNA can become complexed
with polysaccharides and phenolic compounds (Bahloul
and Burkard 1993, Lewinsohn et al. 1994, Salzman et al.
1999), particularly if chaotropic agents (Newbury and
Possingham 1977) or vinyl-pyrrolidone polymers and antioxidants (Salzman et al. 1999) are not present in the extraction buffer. Polysaccharide and phenolic complexes render
the RNA unusable for applications such as reverse transcription and cDNA library construction (Salzman et al.
1999).
Whereas some standard procedures developed for
model plant species, such as Arabidopsis and rice (for example, TRIzol, RNAwiz, RNeasy Plant), are very reliable,
these methods are often inadequate when applied to
leaves of Vitis vinifera. While several reports using recalcitrant plant tissues compared a few methods of RNA isolation (Logemann et al. 1987, Loulakakis et al. 1996, Tesniere and Vayda 1991), the scope of these comparisons
was relatively limited. Logemann et al. (1987) compared six
methods using various combinations of cetyltrimethyl-
1
Department of Biochemistry and Molecular Biology, University of Nevada, Reno, NV 89557; 2current address: Institute of Biotechnology, University of Ankara, Besevler-Ankara, Turkey; and 3current address: Aragene,
The Research Center, King Fahad National Center for Children’s Cancer &
Research, Riyadh, Saudi Arabia.
*Corresponding author [Email: [email protected]; fax 775-784-1650]
Acknowledgments: This work was supported by grants from the American
Viticulture Foundation, the USDA Viticulture Consortium, the National
Science Foundation (NSF) Plant Genome program (#0217653), a graduate
student fellowship to E.A.R. Tattersall from the NSF EPSCoR Integrated
Approaches to Abiotic Stress program (#0132556), and a TUBITAKNATO B1 fellowship to A. Ergul. The Nevada Genomics Center is supported by a grant from the Biomedical Research Infrastructure Network.
The authors wish to thank Kitty Spreeman for lab assistance, Joan Rowe
and Craig Osborne at the Nevada Genomics Center for their assistance with
the Agilent 2100 Bioanalyzer, and Daniel Schachtman for critical reading
of an earlier version of the manuscript.
Manuscript submitted December 2004; revised April, July 2005
Copyright © 2005 by the American Society for Enology and Viticulture.
All rights reserved.
400
Am. J. Enol. Vitic. 56:4 (2005)
Isolating High-Quality RNA from Grapevine Leaves – 401
ammonium bromide (CTAB), cesium chloride (CsCl), and
guanidine for extraction of RNA from potato tubers.
Tesniere and Vayda (1991) applied six methods to grape
berry, five of which included CsCl density centrifugation.
Loulakakis et al. (1996) compared four methods, three including CsCl density centrifugation, on grapevine callus
culture, berry, and leaf. Although many procedures exist, a
more substantial comparison was needed. Therefore, we
compared RNA extraction methods that had been reported
to be successfully used with grapes (Davies and Robinson 1996, Loulakakis et al. 1996, Nassuth et al. 2000, Salzman et al. 1999), as well as other plant species from which
RNA extraction is known to be a challenge (Bahloul and
Burkard 1993, Gehrig et al. 2000, Wang et al. 2000). Fifteen
isolation methods were evaluated and ranked according to
time, cost, quality, and quantity of RNA obtained. Two
methods were found to be the most suitable for a wide
range of molecular biology applications: Tris-lithium chloride (Tris-LiCl) and RNeasy Midi + polyethylene glycol
(PEG).
Materials and Methods
Plant material. Vitis vinifera L. cv. Chardonnay and
Cabernet Sauvignon plants were grown in the field at
Reno, NV (lat. 39o33'N; long. 119o47'W), in pots in greenhouses, in growth chambers (model AR-75L, Percival Scientific, Perry, IA), or hydroponically in greenhouses. Growth
conditions were 16-hr days at 25 to 28°C and 8-hr nights
at 20°C, regardless of location. Plants were grown from
cuttings or rooted cuttings and were one- to two-years
old. Potted plants were grown in Scott’s Metro-Mix 200
(Sun Gro Horticulture, Vancouver, BC). The composition of
the hydroponic solution was according to Gibeaut et al.
(1997). Plants used for RNA isolations were subjected to
a variety of cold (4°C), water deficit, flooding, or heat
(42°C) stresses or were unstressed. Young leaves, young
shoots with leaves, or berries were harvested, immediately
frozen in liquid nitrogen, then stored at -80°C until use
(up to two years). Roots were rinsed to remove soil, blotted dry, and then harvested as above. All RNA methods
were tested on young unstressed leaves. For all methods
except CTAB and CTAB + RNeasy, young stressed leaves
were also used.
RNA protocols. All glassware for RNA extraction was
baked at 180°C for at least 4 hr, whereas plasticware was
either new disposable or treated with 0.1% (v/v) diethylpyrocarbonate (DEPC) water and autoclaved before use.
All solutions used in the RNA extractions, except those
containing Tris, were treated with 0.1% (v/v) DEPC to destroy potentially damaging RNases. Five of the methods
used for RNA isolation were guanidine thiocyanate (Salzman et al. 1999), hot sodium acetate (NaOAc) (Bahloul and
Burkard 1993), TRIzol-PEG (Gehrig et al. 2000), modified
RNeasy method (Nassuth et al. 2000), and Plant RNA Isolation Reagent (Invitrogen, Carlsbad, CA).
Several unpublished modified methods were also used,
including modified hot borate (David Shintani 2001, personal communication) based on Hall et al. (1978). The extraction buffer contained 200 mM sodium borate, 30 mM
[ethylenebis(oxyethylenenitrilo)]-tetraacetic acid, 1% sodium dodecyl sulfate (SDS), 1% (w/v) deoxycholate, 2%
(w/v) polyvinylpyrrolidone (PVP). The solution was autoclaved, and just before use, 10 mM dithiothreitol (DTT)
was added and the extraction buffer heated to 80°C. One
gram of tissue was ground in a liquid-nitrogen-cooled
mortar and placed in a chilled 15-mL tube, then 3.5 mL extraction buffer was added after which 3.5 mL of fresh 1:1
phenol:chloroform was added, and the mixture was vortexed until thoroughly mixed. The liquid phases were
separated by centrifugation at room temperature for 20 min
at 10,000 x g. The aqueous phase was extracted with one
volume of chloroform, adjusted to 2 M lithium chloride
(LiCl) and incubated overnight on ice at 4oC. The precipitate was collected by centrifugation at room temperature
for 20 min at 10,000 x g. The pellet was resuspended in
500 µL DEPC-treated water, after which 50 µL of 2 M potassium acetate pH 5.5 and 1.1 mL of 100% ethanol were
added. The precipitate was pelleted by centrifugation at
10,000 x g for 20 min. The supernatant was completely removed, and the pellet allowed to air dry for 10 min, and
then resuspended in 500 µL DEPC-treated water.
A Tris-LiCl method (Wang et al. 2000) was modified by
adding a phenol:chloroform extraction step. In short, the
extraction buffer consisted of 200 mM Tris-HCl pH 8.5,
1.5% (w/v) lithium dodecyl sulfate, 300 mM Na-EDTA
(ethylenediaminetetraacetic acid), 1% (w/v) sodium deoxycholate, and 1% (v/v) tergitol NP-40. Just before use, 2
mM ATA, 20 mM DTT, 10 mM thiourea, and 2% (w/v)
PVPP (polyvinylpolypyrrolidone) were added. Approximately 1.5 g leaf or 3 g berry or root tissue was ground in
a liquid-nitrogen-cooled mortar and combined with 25 mL
extraction buffer. The mixture was frozen at -80°C for at
least 2 hr. The homogenate was thawed in a 37°C water
bath, and centrifuged for 20 min at 4°C and 5,000 x g. The
supernatant was transferred to fresh tubes, and 0.106 M
NaOAc and 10% (v/v) ethanol added. After 10 min on ice,
this was centrifuged 4°C for 20 min at 5,000 x g. The supernatant was retained and 0.33 M NaOAc and 33% (v/v) isopropanol was added. The mixture was incubated at least 2
hr at -20°C. The tubes were then centrifuged for 30 min at
4°C and 5,000 x g. The pellet was resuspended in 3 mL TE
(10 mM Tris pH 7.5, 1 mM EDTA), and incubated on ice 30
min, then centrifuged for 30 min at 4°C and 5,000 x g. The
supernatant was transferred to fresh tubes, LiCl added to
2.5 M, and incubated on ice at 4°C overnight. The tubes
were centrifuged for 30 min at 4°C and 10,000 x g. The pellet was resuspended in 1.5 mL TE and potassium acetate
was added to 3 M. The mixture was incubated on ice for 3
hr, and centrifuged for 30 min at 4°C and 10,000 x g. The
pellet was resuspended in 1 mL TE, and a single phenol:
chloroform:isoamyl alcohol (25:24:1) extraction performed,
Am. J. Enol. Vitic. 56:4 (2005)
402 – Tattersall et al.
followed by a single chloroform:isoamyl alcohol (24:1) extraction. Ten percent 3.3 M NaOAc and 2 volumes ethanol
were added and incubated at -20°C for at least 2 hr. Tubes
were centrifuged at 4°C and 10,000 x g for 30 min. The
pellets were washed with 500 µL absolute ethanol, dried
on ice, and resuspended in 500 µL DEPC-treated water.
A Tris-LiCl ultracentrifugation method was also performed as described (Loulakakis et al. 1996), except that
CsCl was not added to the RNA solution before ultracentrifugation because it would not dissolve in the solution,
even at a reduced concentration of 0.2 g/mL.
Two modified sodium perchlorate (NaClO 4 ) methods
were tested that were based on the method originally developed by Davies and Robinson (1996). In the first modification, the extraction buffer was changed to contain 0.2
M Tris pH 8.3, and the PEG was omitted. A single TRIzol
extraction was substituted for the phenol:chloroform:
isoamyl alcohol extractions. The second modified method
was from Steve van der Merwe and Johan Burger (2002,
personal communication). The extraction buffer was modified to contain 3 M NaClO4, 0.2 M Tris-HCl pH 8.3, 5% (w/
v) SDS, 8.5% (w/v) PVPP, 2% PEG (w/v), and 1% (v/v) ßmercaptoethanol. One mL of extraction buffer was used for
each 100 mg tissue, and autoclaved Miracloth (Calbiochem, La Jolla, CA) was used to filter the eluate instead
of a glasswool filled syringe. The phenol:chloroform:
isoamyl alcohol extractions were performed repeatedly until the interphase was negligible. At least six phenol:
chloroform:isoamyl alcohol extractions were needed. The
RNA was precipitated using 2 M LiCl. The data from these
two variations of NaClO 4 method were combined in the
analyses, as they have significant similarities.
A CTAB method based on the methods of Jaakola et al.
(2001) and Chang et al. (1993) was modified by Alberto
Iandolino (2003, personal communication). The ground tissue was incubated with extraction buffer for 30 min with
frequent agitation. Two chloroform:isoamyl alcohol extractions were performed. The RNA pellet was resuspended in
DEPC-treated water and a final purification of the RNA
was done using the RNeasy (Qiagen, Valencia, CA) cleanup method.
The RNeasy Midi (Qiagen) + PEG method was performed according to the manufacturer’s instructions for
isolation of RNA from plants, with a few modifications:
3.3% (w/v) PEG (MW 15,000 to 20,000) was added per mL
of buffer RLT before addition of plant material (Gehrig et
al. 2000). Following initial centrifugation, the supernatant
was filtered through Miracloth. Final elution was done
using 150 µL RNase-free water, and the eluate was run
back through the column for the second elution. RNeasy
Mini cleanup was performed as described by the manufacturer, except that no more than 70 µg RNA was applied to
the column.
Reverse transcription PCR. RT-PCR was performed in
a PTC-200 thermocycler (MJ Research, Waltham, MA).
Primers were designed to grape malate dehydrogenase
(GenBank accession # U67426). The sequence of the for-
ward primer was 5’-TCAGCCTGCTTGCCAGTTAC-3’ and
the reverse primer was 5’-CCAAATCTTCTTGCGGGTC-3’.
Reaction conditions were as follows: 1x Invitrogen PCR
buffer (20 mM Tris-HCl [pH 8.4], 50 mM KCl), 3 mM
MgCl 2, 10 mM DTT, 0.5 mM dNTPs, 0.5 µM each primer,
12 units RNAseOUT (Invitrogen), 30 units Superscript II
(Invitrogen), 0.5 units Taq polymerase (Invitrogen), 100 ng
total RNA. The thermocycling conditions were reverse
transcription at 50°C for 30 min, denaturation at 94°C for 2
min, denaturation at 95°C for 1 min, followed by 35 cycles
of denaturation at 94°C for 15 sec, annealing at 50°C for
30 sec, and extension at 72°C for 45 sec, and a final extension at 72°C for 5 min.
RNA analysis. RNA concentration and purity were determined using either a BioWave S2100 (Biochrom, Cambridge, UK) or a Beckman DU530 (Beckman Coulter, Fullerton, CA) spectrophotometer. For purity assessment, the
absorbance for the A260/280 and A260/230 ratios was
taken in water, unless noted otherwise. Formaldehyde-agarose gel electrophoresis was performed according to the
protocol in the Qiagen RNeasy instruction manual. In
brief, 1.2% agarose gels containing ethidium bromide were
prepared. The running buffer consisted of 1 mM EDTA, 5
mM NaOAc, 20 mM 3-(N-morpholino)propanesulfonic acid
(MOPS), adjusted to pH 7.0 with NaOH, and 246 mM formaldehyde was added. The 5x RNA loading dye consisted
of 80 mM MOPS, 20 mM NaOAc, 8 mM EDTA, 886 mM
formaldehyde, 20% (v/v) glycerol, 31% (v/v) formamide,
0.16% (v/v) saturated bromophenol blue. One microgram
of total RNA was loaded, except as indicated otherwise.
Before loading, RNA with loading dye was heated at 65°C
for 4 min, then chilled on ice. The gel was run at 7 V/cm.
Gels were imaged using the GelDoc 2000 gel documentation system and QuantityOne image analysis software version 4 (Bio-Rad, Hercules, CA). Agilent 2100 Bioanalyzer
RNA LabChip assays were performed following the
manufacturer’s instructions (Agilent Technologies, Palo
Alto, CA). All statistical analyses were performed using
JMP (version 5, SAS Institute, Cary, NC).
Results and Discussion
RNA yield. There were large differences in the amount
of RNA extracted per gram of tissue depending on the
protocol used (Table 1). For example, the CTAB and
TRIzol-PEG methods gave the highest mean raw yields in
micrograms RNA per gram fresh weight (846 and 821 µg/g
FW, respectively). In contrast, the modified RNeasy
method gave the lowest mean yield (13 µg/g FW). The
Tris-LiCl method gave a moderate yield of over 300 µg/g
FW. Yields were significantly reduced by the addition of
an RNeasy cleanup step (Tables 1 and 2). Mean yield from
all methods for stressed tissue was 399 ± 34 µg/g FW and
was not significantly different from unstressed tissue (392
± 39 µg/g FW, p = 0.89).
RNA quality. RNA quality was assessed by five methods: A260/280, A260/230, appearance on a denaturing agarose gel, ability to produce RT-PCR products, and Agilent
Am. J. Enol. Vitic. 56:4 (2005)
Isolating High-Quality RNA from Grapevine Leaves – 403
2100 Bioanalyzer RNA LabChip assay. A260/280 ratios indicate the level of protein contamination, based on the
principle that nucleic acids display an absorbance optimum at 260 nm, whereas proteins (primarily aromatic amino
acids) display an absorbance optimum at 280 nm (Winfrey
et al. 1997). After running on a denaturing gel, high-quality RNA shows two or more sharp rRNA bands with little
smearing, depending on whether the gel is run long
enough to separate the 16S from
the 18S and the 23S from the 25S
rRNA. The modified RNeasy, RNeasy
Table 1 Comparison of yield and quality from various methods of RNA isolation from
grapevine young leaves as assessed using a spectrophotometer.
Midi + PEG, CTAB + RNeasy, and
Tris-LiCl ultracentrifugation methAbsorbance
ods gave RNA with very low
RNA yield
amounts of protein contamination
Method
na
(µg/g FW) b,c
260/280b,d
260/230b,d
(Table 1). In contrast, RNA isolated
CTAB
2
846 ± 68AB
1.45 ± 0.35 EF
1.67 ± 0.06DEFGH
by the hot borate and CTAB methTRIzol-PEG
9
821 ± 242A
1.61 ± 0.18 EF
1.70 ± 0.05EFG
ods showed significantly more protein contamination, as indicated by
Hot borate
10
665 ± 309 A
1.49 ± 0.06F
0.95 ± 0.21I
the lower A260/280 ratios.
Plant RNA Isolation
46
624 ± 77A
1.74 ± 0.02DEF
1.22 ± 0.02HI
Reagent
A260/230 ratios are used to assess the level of contamination by
Tris-LiCl
75
387 ± 23BC
1.80 ± 0.01DE
2.49 ± 0.03BC
polysaccharides and polyphenols
Tris-LiCl + RNeasy
42
302 ± 15CD
1.70 ± 0.01 EF
2.56 ± 0.05B
(Loulakakis et al. 1996, Schultz et al.
CD
DEF
EF
Plant RNA Isolation
23
299 ± 20
1.75 ± 0.04
1.73 ± 0.07
Reagent + RNeasy
1994). A ratio of 2.5 is considered
free of contamination. The modified
Tris-LiCl ultracentrifugation
10
228 ± 87CDE
2.13 ± 0.11BC
2.40 ± 0.08BC
RNeasy and hot borate methods
CDE
F
GHI
Tris-LiCl-Wang
8
159 ± 72
1.54 ± 0.05
1.28 ± 0.10
yielded RNA that was significantly
Guanidine thiocyanate
10
97 ± 63CDE
1.79 ± 0.39CDEF
1.47 ± 0.28FGH
contaminated with polysaccharides
CTAB + RNeasy
6
85 ± 10DE
2.00 ± 0.06CD
1.91 ± 0.11DE
(Table 1). In contrast, the RNeasy
RNeasy Midi + PEG
4
54 ± 3.0DE
2.76 ± 0.06 A
4.21 ± 0.49A
Midi + PEG, Tris-LiCl ultracentriNaClO 4
11
40 ± 8.0 E
1.87 ± 0.13CDE
2.23 ± 0.23CD
fugation, Tris-LiCl, and Tris-LiCl +
RNeasy methods yielded RNA with
Hot NaOAc
6
26 ± 7.0DE
1.61 ± 0.08 EF
1.27 ± 0.03FGHI
very little polysaccharide.
E
AB
J
Modified RNeasy
9
13 ± 3.0
2.41 ± 0.45
0.21 ± 0.04
The yield and quality of RNA
a
n: number of extractions.
from the Tris-LiCl method was imbMeans ± SE; superscript uppercase letters indicate p ≤ 0.05 significance level based on least
proved significantly by the addition
square means student’s t test.
c
of a phenol:chloroform step as comYield is given in µg RNA per gram fresh weight.
d
pared to the original Tris-LiClA260/280 and A260/230 taken in water.
Wang (2000) protocol (Table 1). The
Tris-LiCl ultracentrifugation method (Loulakakis et al.
1996) yielded RNA that had excellent purity and amount
Table 2 Comparison of leaf RNA before and after
based on spectrophotometric measurements, but 6 of 10
RNeasy cleanup isolated by Plant RNA Isolation Reagent
and Tris-LiCl methods.
samples showed significant to severe degradation upon
denaturing agarose gel electrophoresis.
Isolation
Most of the methods for which RNA could be seen on
Solvent
Reagent a,b
Tris-LiCla,c
a gel had a clear band pattern, although many had some
Before cleanup
smearing (Figure 1A), indicative of RNA degradation.
A260/280
Water
1.72 ± 0.03
1.82 ± 0.03
These patterns were reproducible and are seen more
A260/280
Tris
1.80 ± 0.01
2.05 ± 0.02
clearly in the gel image and electropherograms from the
A260/230
Water
1.30 ± 0.04
2.65 ± 0.04
Agilent 2100 Bioanalyzer (Figure 1B, C).
µg/g FW
Water
624 ± 77
387 ± 23
RNA preparations were further tested for quality using
RT-PCR reactions, performed in triplicate. The RNA from
After RNeasy
NaClO4, TRIzol-PEG, RNeasy Midi + PEG, Tris-LiCl, and
A260/280
Water
1.68 ± 0.02
1.70 ± 0.01
A260/280
Tris
2.13 ± 0.01
2.18 ± 0.01
Tris-LiCl + RNeasy methods consistently resulted in amA260/230
Water
1.81 ± 0.11
2.66 ± 0.07
plification (Table 3, Figure 2), while the CTAB, guanidine
µg/g FW
Water
299 ± 20
302 ± 15
thiocyanate, hot NaOAc, and Tris-LiCl ultracentrifugation
methods consistently gave no amplification.
aData based on a subset of what is listed in Table 1 for which meaCost. Approximate hands-on time required for each
sures were taken in both water and 10 mM tris pH 7.5.
b24 extractions.
method and cost per µg RNA were analyzed. Hands-on
c12 extractions.
time included incubations or centrifugations of less than
Am. J. Enol. Vitic. 56:4 (2005)
404 – Tattersall et al.
Figure 1 Leaf RNA quality as assessed by denaturing agarose gel electrophoresis and Agilent Bioanalyzer 2100. (A) Agarose gel image stained with
ethidium bromide. (B) Agilent gel image. (C) Agilent electropherograms; note that scales differ. Lane 1, Plant RNA Isolation Reagent; 2, Plant RNA
Isolation Reagent + RNeasy; 3a, CTAB; 3, CTAB + RNeasy; 4, guanidine thiocyanate; 5, hot NaOAc; 6, NaClO4; 7, hot borate; 8, modified RNeasy;
9, TRIzol-PEG; 10, Tris-LiCl; 11, Tris-LiCl + RNeasy; 12, Tris-LiCl ultracentrifugation; 13, RNeasy Midi + PEG. L, ladder.
one hour: longer periods were not included, as researchers may perform other tasks. The labor involved in preparing solutions was not included in the cost estimates. Total
cost per µg RNA isolated was based on mean yields, reagent costs, labor at $8.00/hr, and the cost of single-use
tubes, filters, or columns. Generally, the higher the RNA
yield per gram of tissue, the lower the cost per µg, regardless of the cost of reagents or the labor involved. Some
methods can be scaled up, which gives a savings in labor
costs per µg, while scaling down can increase the cost
per µg RNA. The Tris-LiCl ultracentrifugation, CTAB,
Plant RNA Isolation Reagent, and Tris-LiCl methods were
the most economical (Table 3), and the guanidine thiocyanate and modified RNeasy methods were the most costly
because of low yields and/or small scale.
Application of RNA methods to roots and berries. We
have used the Tris-LiCl + RNeasy method extensively, and
it was used to isolate RNA for the leaf, root, and berry
cDNA libraries that our group has prepared for EST sequencing. When run on a gel, four rRNA (16S, 18S, 23S,
and 25S) bands can be seen for Tris-LiCl + RNeasy isolated RNA. Although this method is relatively time-consuming, it consistently yields a good quantity of quality
RNA, is economical, and does not require specialized
equipment. We have stored Tris-LiCl + RNeasy isolated
RNA at -80°C for more than six months without noticeable
degradation in quality, and even after four freeze-thaw
cycles it is still competent for RT-PCR. Furthermore, we
have performed high-quality global transcript analysis with
Tris-LiCl + RNeasy isolated RNA using the Vitis Affymetrix
Am. J. Enol. Vitic. 56:4 (2005)
Isolating High-Quality RNA from Grapevine Leaves – 405
GeneChip (www.affymetrix.com). The average
coefficient of variation for six technical replicates from young shoots of unstressed
RT-PCR
plants was low (16%), indicating that the
RNA a
Hands-on
Elapsed
success
b
c
quality of the RNA was good.
rate
Method
($/µg)
time (hr)
time
In a few cases, we have had some deTris-LiCl ultracentrifugation
0.05
6
2-3 days
0/3
graded RNA from berry extractions using the
CTAB
0.05
5
2 days
0/3
Tris-LiCl method. The RNeasy Midi + PEG
Plant RNA Isolation
0.06
3.5
3.5 hr
1/3
method was tested for berry RNA isolation,
Reagent
and it gave consistently good quality and
Tris-LiCl
0.07
10
2-4 days
3/3
adequate yield in a much shorter time than
Tris-LiCl + RNeasy
0.13
11
3-4 days
3/3
the Tris-LiCl + RNeasy method. This method
TRIzol-PEG
0.15
3
4 hr
3/3
was also tested on leaves and roots (Tables
Plant RNA Isolation
0.22
4.5
4.5 hr
1/3
1 and 3, Figure 2). It seems to be very effecReagent + RNeasy
tive at removing polysaccharides and phenoHot borate
0.23
3.5
2 days
0/3
lics from leaves and berries; the absorbance
CTAB + RNeasy
0.54
5
2 days
0/3
at 230 nm was nearly undetectable for many
samples, giving unusually high A260/230 raNaClO 4
0.80
3.5
3.5 hr
3/3
tios (Table 1). RNeasy Midi + PEG RNA from
Hot NaOAc
0.93
2.5
2 days
0/3
leaves or berries was RT-PCR competent (3/3
RNeasy Midi + PEG
0.97
4
4 hr
3/3
reactions), but the RNA from roots failed to
Guanidine thiocyanate
4.62
3.5
2-3 days
0/3
amplify by RT-PCR.
Modified RNeasy
9.92
1
1 hr
0/3
The Tris-LiCl method was also applied
aCost includes hands-on time at $8.00/hr.
successfully both to roots and to berries at
bRanges in elapsed time depend on choice of minimum precipitation time or overvarious developmental stages (from preverainight.
son to over-ripeness). The mean yield for
c Number of reactions where product was obtained over number of reactions atroots was 20 mg/g FW and for berries was
tempted.
40 µg/g FW (Table 4). With a Tris-borate
method, Franke et al. (1995) reported that
good-quality RNA was obtained from berries at stages up
to the beginning of veraison, but not from postveraison
berries. The average yield for five preveraison stages of
berries was 26 µg/g FW. Using an extraction protocol
Figure 2 Leaf RNA quality as assessed by RT-PCR with grape malate
dehydrogenase primers (one representative gel out of three). Lane 1,
similar to the Tris-LiCl ultracentrifugation and Tris-LiCl
Plant RNA Isolation Reagent; 2, Plant RNA Isolation Reagent + RNeasy;
methods, Tesniere and Vayda (1991) reported yields from
3, CTAB; 4, CTAB + RNeasy; 5, guanidine thiocyanate; 6, hot NaOAc;
berries of 18 µg/g FW, based on at least three replicates.
7, NaClO4; 8, hot borate; 9, modified RNeasy; 10, TRIzol-PEG; 11, TrisThese yields are half of what we obtained using the TrisLiCl; 12, Tris-LiCl + RNeasy; 13, Tris-LiCl ultracentrifugation; 14, RNeasy
Midi + PEG.
LiCl method (40 µg/g FW) and their method requires an ultracentrifugation step. A yield of 30 µg/g FW was obtained from berries using the RNeasy Midi + PEG method.
Table 3 RT-PCR success rate and cost and time required for
each extraction method based on leaf yields.
Table 4 Mean (±SE) yield and quality for Tris-LiCl and RNeasy
Midi + PEG isolated root, berry, and leaf RNA as assessed
spectrophotometrically.
Method
Organ
na
Yield
(µg/g FW)
A260/280
A260/230
Tris-LiCl
Root
55
20 ± 5
1.88 ± 0.05
2.07 ± 0.20
Tris-LiCl
Berry
27
40 ± 3
1.65 ± 0.03
1.80 ± 0.10
Tris-LiCl
Leaf
75
387 ± 23
1.80 ± 0.01
2.49 ± 0.03
RNeasy Midi
+ PEG
Root
6
57 ± 6
1.50 ± 0.04
1.01 ± 0.02
RNeasy Midi
+ PEG
Berry
36
30 ± 1
2.58 ± 0.05
18.5 ± 3.3b
RNeasy Midi
+ PEG
Leaf
6
54 ± 3
2.76 ± 0.06
4.21 ± 0.49b
a
b
n: number of isolations.
A230 was near zero for some samples, resulting in unusually high
ratios.
Conclusions
The choice of method depends upon the application in
which the RNA will be used. Yield, A260/280, and A260/
230 ratios are not necessarily good indicators of RT-PCR–
competent RNA. RNA should be checked for degradation. For RT-PCR, only four extraction protocols gave
consistent product: NaClO 4, TRIzol-PEG, RNeasy Midi +
PEG, and Tris-LiCl (with or without RNeasy). The TRIzolPEG extraction had significant RNA degradation, making
quantification of RT-PCR questionable. Sodium perchlorate
extractions had slight RNA degradation, low extraction efficiency, and cost substantially more than the Tris-LiCl
method. The modified Tris-LiCl + RNeasy method was a
good choice for large quantities of RNA because it provided a good yield and excellent quality for low cost; it
Am. J. Enol. Vitic. 56:4 (2005)
406 – Tattersall et al.
also provided high-quality and consistent results for
shoots and leaves for both RT-PCR and transcript profiling
using oligonucleotide-based microarrays. The RNeasy
Midi + PEG method gave lower yields and was significantly more expensive than Tris-LiCl + RNeasy, but it was
faster and provided high-quality undegraded RNA. It was
also less prone to error but was unsuitable for obtaining
RT-PCR–competent RNA from roots. For small quantities
of RNA from leaves or berries, such as for RT-PCR or
microarrrays, the RNeasy Midi + PEG method provided
ease of use and excellent quality.
John, M.E. 1992. An efficient method for isolation of RNA and DNA
from plants containing polyphenolics. Nucleic Acids Res. 20:2381.
Literature Cited
Loulakakis, K.A., K.A. Roubelakis-Angelakis, and A.K. Kanellis.
1996. Isolation of functional RNA from grapevine tissues poor in
nucleic acid content. Am. J. Enol. Vitic. 47:181-185.
Bahloul, M., and G. Burkard. 1993. An improved method for the
isolation of total RNA from spruce tissues. Plant Mol. Biol. Rep.
11:212-215.
Chang, S., J. Puryear, and J. Cairney. 1993. A simple and efficient
method for isolating RNA from pine trees. Plant Mol. Biol. Rep.
11:113-116.
Davies, C., and S.P. Robinson. 1996. Sugar accumulation in grape
berries: Cloning of two putative vacuolar invertase cDNAs and
their expression in grapevine tissues. Plant Physiol. 111:275-283.
Franke, K.E., Y. Liu, and D.O. Adams. 1995. Yield and quality of
RNA from grape berries at different developmental stages. Am.
J. Enol. Vitic. 46:315-318.
Gehrig, H.H., K. Winter, J. Cushman, A. Borland, and T. Taybi.
2000. An improved RNA isolation method for succulent plant
species rich in polyphenols and polysaccharides. Plant Mol. Biol.
Rep. 18:369-376.
Gibeaut, D.M., J. Hulett, G.R. Cramer, and J.R. Seemann. 1997.
Maximal biomass of Arabidopsis thaliana using a simple, lowmaintenance hydroponic method and favorable environmental conditions. Plant Physiol. 115:317-319.
Hall, T.C., Y. Ma, B.U. Buchbinder, J.W. Pyne, S.M. Sun, and
F.A. Bliss. 1978. Messenger RNA for G1 protein of french bean
seeds: Cell-free translation and product characterization. PNAS.
75:3196-3200.
Jaakola, L., A.M. Pirttila, M. Halonen, and A. Hohtola. 2001. Isolation of high-quality RNA from bilberry (Vaccinium myrtillus L.)
fruit. Mol. Biotech. 19:201-203.
Katterman, F.R., and V.I. Shattuck. 1983. An effective method of
DNA isolation from the mature leaves of Gossypium species that
contain large amounts of phenolic terpenoids and tannins. Prep.
Biochem. 13:347-359.
Lewinsohn, E., C.L. Steele, and R. Croteau. 1994. Simple isolation
of functional RNA from woody stems of gymnosperms. Plant Mol.
Biol. Rep. 12:20-25.
Logemann, J., J. Schell, and L. Willmitzer. 1987. Improved method
for the isolation of RNA from plant tissues. Anal. Biochem. 163:1620.
Nassuth, A., E. Pollari, K. Helmeczy, S. Stewart, and S.A. Kolfalvi.
2000. Improved RNA extraction and one-tube RT-PCR assay for
simultaneous detection of control plant RNA plus several viruses
in plant extracts. J. Virol. Meth. 90:37-49.
Newbury, H.J., and J.V. Possingham. 1977. Factors affecting the
extraction of ribonucleic acid from plant tissues containing interfering phenolic compounds. Plant Physiol. 60:543-547.
Salzman, R.A., T. Fujita, K. Zhu-Salzman, P.M. Hasegawa, and R.A.
Bressan. 1999. An improved RNA isolation method for plant tissues containing high levels of phenolic compounds or carbohydrates.
Plant Mol. Biol. Rep. 17:11-17.
Sambrook, J., E.F. Fritsch, and T. Maniatis. 1989. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press,
Plainview, NY.
Schultz, D.J., R. Craig, D.L. Cox-Foster, R.O. Mumma, and J.I.
Medford. 1994. RNA isolation from recalcitrant plant tissue. Plant
Mol. Biol. Rep. 12:310-316.
Tesniere, C., and M.E. Vayda. 1991. Method for the isolation of
high-quality RNA from grape berry tissues without contaminating tannins or carbohydrates. Plant Mol. Biol. Rep. 9:242-251.
Wang, S.X., W. Hunter, and A. Plant. 2000. Isolation and purification of functional total RNA from woody branches and needles
of sitka and white spruce. BioTechniques 28:292-296.
Winfrey, M.R., M.A. Rott, and A.T. Wortman. 1997. Unraveling
DNA: Molecular Biology for the Laboratory. Prentice-Hall, Upper Saddle River, NJ.
Am. J. Enol. Vitic. 56:4 (2005)
Isolating High-Quality RNA from Grapevine Leaves – 407
Am. J. Enol. Vitic. 56:4 (2005)