|Home | About | Journals | Submit | Contact Us | Français|
Variability of leaf structure and presence of secondary metabolites in mature leaf tissue present a challenge for reliable DNA extraction from Osmanthus species and cultivars. The objective of this study was to develop a universal rapid, effective, and cost-efficient method of DNA isolation for Osmanthus mature leaf tissue. Four different methods were used to isolate DNA from 8 cultivars of Osmanthus. Absorbance spectra, DNA concentration, appearance on agarose gel, and performance in PCR were used to analyze quality, quantity, and integrity of isolated DNA. Methods were ranked in order, based on total quantity, quality, and performance points as the following: 1) solid-phase extraction (SPE), 2) modified alkaline lysis (SDS), 3) cetyltrimethylammonium bromide (CTAB) with chloroform (CHL), and 4) CTAB with phenol/chloroform (PHE). Total DNA, isolated via SPE, showed the least contamination but the lowest mean quantity (9.6 ± 3.4 μg) and highest cost. The highest quantity of DNA was isolated via SDS (117 ± 54.1 μg). SPE and SDS resolved the most individuals on agarose gel, whereas the 2 CTAB methods had poorly resolved gels. All methods except PHE performed well in PCR. Additions to the modified alkaline lysis method increased A260:A230 by up to 59% without affecting yield. With the use of SDS, an average of 1000 μg/g DNA was isolated from fresh leaf tissue of 18 samples in ~1.5 h at a cost of 0.74 U.S. dollars (USD)/sample. We recommend improved alkaline lysis as a rapid, effective, and cost-efficient method of isolating DNA from Osmanthus species.
Osmanthus (Oleaceae) is a genus of evergreen trees and shrubs, comprised of ~35 species, distributed primarily in temperate to subtropical parts of eastern Asia.1–3 The most popular species, Osmanthus fragrans (flowering tea olive) is a top 10 medicinal plant of China, includes at least 166 cultivars, and is widely available in the commercial trade. At least 10 species of Osmanthus are cultivated.3 Osmanthus heterophyllus, Osmanthus x fortunei, and Osmanthus delavayi have several popular horticultural forms that vary widely in size, habit, and leaf form. The diversity of plant form and type, combined with the wide geographic range covered by the genus and its close relationship to other important genera, makes it attractive for genetic improvement.
As a nonmodel ornamental plant genus, there are few molecular tools available for Osmanthus. Interest in species improvement currently centers on color, fragrance, and chemical profile of O. fragrans flowers, with increased cold tolerance, increased tolerance to drought and disease, and improved landscape form desired for other commercially available species. Initial progress in advancing a molecular toolkit for Osmanthus includes the production of microsatellites in O. fragrans through enriched library development and transcriptome sequencing and Osmanthus phylogenies and genetic relationships inferred from chloroplast and nuclear markers.4–9 These studies, which collectively include 22 Osmanthus species, report genomic DNA isolation by the CTAB method of Doyle and Doyle.10 None of the studies mention amount or purity of isolated DNA; cost, time, or expense of the isolation procedure; or relative ease or difficulty of the species examined.
CTAB, a cationic surfactant, is the de facto detergent of choice to solubilize cell membranes and denature cell lysate proteins during DNA isolation.10, 11 CTAB lysis is often followed by organic extraction by use of alcohol-saturated chloroform or chloroform with phenol. Two alternatives to organic extraction include solid-phase extraction (SPE, the basis of many commercial DNA isolation kits) and alkaline lysis. Solid-phase DNA extraction uses a spin column operated under centrifugal force. Silica matrices, glass particles, diatomaceous earth, and anion-exchange carriers are often used as solid support in SPE.12–15 Alkaline lysis uses the ionic detergent sodium dodecyl sulfate and sodium hydroxide or salt in the lysis buffer to solubilize cell membranes and denature proteins in the cell lysate.16, 17 Both SPE and alkaline lysis have proven to yield high-quality DNA from recalcitrant plant species.18–21
To extend the molecular toolkit into less-visible Osmanthus species and to develop more sophisticated tools, such as genetic maps, quantitative trait loci, and genotype- or marker-assisted selection, a rapid, cost-effective method of DNA isolation is required that can isolate large amounts of high-quality DNA from diverse species by use of mature leaves. The objective of this study was to compare 4 different methods of DNA isolation across 4 species of Osmanthus, based on amount and purity of isolated DNA, performance in PCR, and the time and cost associated with each method.
Eight cultivars of Osmanthus, representing 4 species, were used for analysis. O. fragrans "Beni Kin Mokusei" and "Fudingzhu," Osmanthus armatus "Jim Porter" and "Longwood," O. heterophyllus "Goshiki" and "Head-Lee Fastigate," and O. fortunei "San Jose" and "Fruitlandii" were chosen based on their commercial availability and diversity of leaf type. Plants were purchased as rooted cuttings from Nurseries Caroliniana (North Augusta, SC, USA) and maintained in 3-gallon pots, except for O. fragrans, which were maintained in 7-gallon pots. Hydrangea macrophylla "Veitchii" was used as an outgroup; H. macrophylla and O. fragrans were considered “easy” species for DNA isolation. Two replicates of each sample were used in DNA isolation (n = 18 samples/method).
For all samples, mature leaves were collected, and 100 mg tissue, inclusive of the leaf tip, was placed in a 2 ml screw-top microcentrifuge tube containing 2, 6 mm Pyrex beads (Corning, Corning, NY, USA). Lysis buffer (800 μl–1 ml; specific to treatment) was added, and tubes were subjected to 3 cycles of shaking at 6 m/s for 40 s in a FastPrep FP120 Cell Disruptor (Qbiogene, Carlsbad, CA, USA).
The Qiagen DNeasy Plant Mini Kit uses spin columns with silica gel membranes. Nucleic acids are adsorbed to the silica gel membrane in the presence of chaotropic salts, whereas polysaccharides and proteins do not adsorb and are removed. After a wash step, pure nucleic acids are eluted under low- or no-salt conditions in small volumes.15
The Qiagen DNeasy Plant Mini Kit was used according to the manufacturer’s instructions, starting from step 8 of the plant tissue mini protocol.15 Samples were disrupted in 800 μl lysis buffer (Qiagen Buffer AP1) with 8 μl 100 mg/ml RNAse A and incubated at 65°C for 10 min. Subsequent steps were performed as per the protocol. Samples were eluted twice with 100 μl elution buffer (Qiagen Buffer AE) for a total of 200 μl product.
This method was adapted from Doyle and Doyle.10 Samples were disrupted in 1 ml CTAB lysis buffer [2% CTAB, 1.4 M NaCl, 100 mM Tris-HCl, 20 mM EDTA (TE), pH 8]. The sample was incubated at 65°C for 1 h with occasional mixing, followed by centrifugation for 5 min at 16,000 g. Supernatant (500 μl) was transferred to a new tube; 5 μl, 10 mg/ml RNase A was added to each sample; and samples were incubated at 37°C for 15 min. One volume (505 μl) 24:1 chloroform:isoamyl alcohol was added to each sample. Samples were inverted for 2 min and centrifuged for 5 min at 16,000 g. The aqueous phase (300 μl) was removed to a 2 ml tube, and 0.5 vol 5 M NaCl (150 μl), followed by 3 vol ice-cold absolute ethanol (1350 μl), was added to each sample. Samples were inverted for 30 s, incubated at −20°C for 1 h, and centrifuged for 10 min at 16,000 g. The supernatant was removed and the DNA pellet washed twice with 300 μl ice-cold 70% ethanol. The ethanol was removed by pipetting and the pellet allowed to air dry for 20 min. DNA pellets were resuspended in 200 μl TE.
Phenol can be added to chloroform to aid in the removal of proteins from cell lysate during DNA extraction. Phenol causes proteins to denature and migrate to a nonaqueous phase, which can be removed without disturbing the DNA-containing aqueous phase.
Samples were prepared and treated as in Method 2 (above) except that 505 μl 25:24:1 phenol:chloroform:isoamyl alcohol was used during the organic extraction step. DNA was precipitated by adding 0.5 vol 7.5 M ammonium acetate (150 μl), followed by 2.5 vol ice-cold absolute ethanol (1125 μl) to each sample. Incubation at −20°C and subsequent steps were performed as for Method 2.
This method was adapted from a modification of the Edwards method of alkaline lysis.17, 21 Samples were disrupted in 1 ml SDS lysis buffer (0.5% sodium dodecyl sulfate, 200 mM NaCl, 25 mM EDTA, 200 mM Tris-HCl, pH 8). Samples were incubated at 95°C for 10 min and then centrifuged for 10 min at 2000 rpm. The supernatant (500 μl) was transferred to a 1.5 ml microcentrifuge tube and centrifuged again for 10 min at 2000 rpm. The supernatant (400 μl) was transferred to a new 1.5 ml microcentrifuge tube, and 1 vol ice-cold isopropanol was added. Samples were inverted for 30 s, incubated at room temperature for 10 min, and centrifuged for 10 min at 16,000 g. The supernatant was discarded and the pellet allowed to air dry for 20 min. DNA pellets were resuspended in 200 μl TE, 4 μl 10 mg/ml RNase was added, and samples were incubated at 37°C for 15 min.
A separate experiment was carried out to examine the impact of additional steps and/or reagents on quantity and quality of DNA isolated via modified alkaline lysis. Replicate samples of O. fragrans Fudingzhu and O. armatus Jim Porter were prepared as for Method 4 (above). Four total samples were used in each of 5 treatments: no changes from Method 4 (Control), 2 ethanol washes (Wash), addition of 1 μl 20 mg/ml glycogen to each sample during DNA precipitation (GLY), addition of 20 μl 2-ME to each sample during sample disruption (2ME), and addition of Proteinase K (final concentration, 20 mg/ml) to the lysis buffer (PK). After the Control samples were quantified, the DNA pellet was discarded, and the remaining DNA-saturated buffer was quantified (DP). The GLY, 2ME, and PK treatments also included Wash, and the DNA pellets were discarded. Five additional samples underwent a Wash treatment but were incubated at 4°C overnight before the DNA pellets were discarded. Samples were quantified before and after overnight incubation.
DNA was evaluated by amplifying the 1600 bp chloroplast trnS-psbC gene region for each cultivar [primer 1: 5′-GGT TCG AAT CCC TCT CTC TC-3′; primer 2 (reverse): 5′-GGT CGT GAC CAA GAA ACC AC-3′]. Each 25 μl PCR was carried out by use of the Qiagen Core PCR Kit with 1.5 mM MgCl2, 100 mM each dNTP, 0.2 μM each primer, 1 unit Taq DNA polymerase, and 50 ng template DNA. Cycling conditions were as follows: 3 min initial denaturation at 94°C, 25 cycles of denaturation for 30 s at 94°C, annealing for 30 s at 52°C, and extension for 1 min at 70°C and a final extension for 10 min at 70°C. Genomic DNA was electrophoresed on 1.0% wt/vol agarose with 0.2 μg/ml ethidium bromide in Tris-acetate buffer at 10 V/cm for 1 h and visualized by UV fluorescence. PCR products were electrophoretically separated on 2.0% wt/vol agarose with 0.2 μg/ml ethidium bromide in Tris-acetate buffer at 10 V/cm for 1 h and visualized by UV fluorescence.
Genomic DNA was quantified by use of the NanoDrop 2000 spectrophotometer (Life Technologies, Thermo Fisher Scientific, Indianapolis, IN, USA), which measures nucleic acid sample concentrations based on sample absorbance at 260 nm, the wavelength-dependent extinction coefficient (50 ng cm/μl for DNA), and the pathlength between optical fibers embedded in the measurement surfaces. For each sample, DNA concentration (ng/μl) and sample absorbances at 230, 260, and 280 nm were recorded. DNA concentrations were converted into total DNA recovered, and 260:230 and 260:280 absorbance ratios were calculated. All statistical analysis was performed by use of SAS statistical analysis software (v.9.3; SAS Institute, Cary, NC, USA). Nested 2-way ANOVA was used to partition variance in DNA concentration and quality into factors attributable to species, cultivar, and method. Cultivar was nested within species, as each cultivar is unique to a single species. Tukey’s studentized range test was used to separate means at the 0.05 level of significance.
Extraction methods were compared by assigning points to quantity, quality, and performance variables. Each sample was given a score from 0 to 3, where 0 was the “worst” score, and 3 was the “best” score (Table 1). Each method consisted of 18 samples; thus, each method could receive a maximum of 54 points/variable and 270 points total. Points for each cultivar and species were tabulated by use of the same method. Each cultivar was represented by 8 total samples across all methods; thus, each cultivar could receive a maximum of 24 points/variable and 120 total points. A low score is interpreted as a cultivar or species from which DNA isolation is difficult, whereas a relatively high score is interpreted as a cultivar or species from which DNA isolation is easier.
Four different DNA isolation methods (SPE, CHL, PHE, and SDS) were used to isolate genomic DNA from 8 cultivars of Osmanthus and 1 cultivar of H. macrophylla. Methods were assigned points based on quantity, quality, and performance of isolated DNA (see Supplemental Table 1 for individual sample-point assignments). SPE scored the highest total points (196 out of 270), followed by SDS (172 out of 270; Table 2). The 2 traditional CTAB methods (CHL and PHE) scored 161 and 157 out of 270 points. Cost/sample varied >5-fold among methods (see Supplemental Tables 2–4 for a detailed cost analysis). SPE cost 5 USD/sample, whereas SDS cost 0.74 USD/sample. CHL and PHE averaged ~1.50 USD/sample. CHL and PHE took the most time, requiring 3.5 h for DNA isolation from 18 samples. SPE required 2.5 h, and SDS required 1.5 h for 18 samples. Points were also assigned to each cultivar and species based on quantity, quality, and performance of isolated DNA (Table 3). Species ranked in order from high to low score as follows: O. fortunei, H. macrophylla, O. fragrans, O. heterophyllus, and O. armatus. O. fortunei Fruitlandii was the highest scoring cultivar tested, whereas O. armatus Jim Porter was the lowest scoring, based on total quantity, quality, and performance points averaged across all methods.
Total DNA recovered was highly variable and ranged between 9.6 and 117 μg, with an overall mean of 43.0 ± 49.9 μg or ~430 μg/g fresh leaf tissue. SDS yielded significantly more DNA than the other 3 methods (117 ± 54.1 μg; Fig. 1). SPE yielded the least (9.6 ± 3.4 μg) but not significantly less than the CHL or PHE (27.4 ± 22.4 μg and 17.9 ± 14.9 μg, respectively). Within each method, DNA yields were highly variable except for SPE, where yields were more uniform. The total DNA recovered was significantly influenced by the cultivar, method, and interaction between cultivar and method (Table 4). Most of the variation was accounted for by method, as SDS yielded significantly more than any other method. However, the uniformly low yields of O. armatus Jim Porter led to a significant source of cultivar variation, as well as interaction variation. By total points (Table 2), methods were ranked by quantity of DNA in the following order: SDS > CHL > PHE > SPE.
A260:A280 and A260:A230 ratios ranged between 1.7 and 2.0 and 0.86 and 2.4, respectively. Overall mean A260:A280 was 1.9 ± 0.13, whereas overall mean A260:A230 was 1.7 ± 0.63. The A260:A280 ratio was significantly influenced by the cultivar, method, and interaction between cultivar and method (Table 4). SPE and SDS had the lowest A260:A280 ratios, and their 95% confidence limits included 1.8, which is considered the standard for “pure” DNA. CHL and PHE had mean A260:A280 ratios higher than 1.8, which often happens when DNA concentration is very high and is not necessarily indicative of contamination. After diluting of samples to 50 ng/μl and requantifying the DNA, A260:A280 95% confidence limits for all methods included 1.8, although ranking of methods based on mean did not change. By total points (Table 2), methods were ranked by quality of DNA, as assessed by the A260:A280 ratio in the following order: SPE > PHE > CHL > SDS.
The A260:A230 ratio was also influenced significantly by the cultivar, method, and interaction between cultivar and method (Table 4), although method accounted for a much larger share of the variation relative to the other variables. DNA isolated by use of SPE had the highest mean A260:A230 value (2.4 ± 0.11), whereas SDS had the lowest mean A260:A230 (0.86 ± 0.28), well under the target of 2.0–2.2. Mean absorbance ratios for SPE and PHE DNA had 95% confidence limits that contained the A260:A230 target purity range of 2.0–2.2. O. fortunei DNA had the highest A260:A230 ratio, whereas O. armatus and H. macrophylla had the lowest ratios. By total points (Table 2), methods were ranked by quality of DNA, as assessed by the A260:A230 ratio in the following order: SPE > PHE > CHL > SDS.
Genomic DNA was electrophoresed on 1.0% wt/vol agarose with 0.2 μg/ml ethidium bromide in Tris-acetate buffer at 10 V/cm for 1 h and visualized by UV fluorescence. Undiluted genomic DNA was used to confirm quantity and estimate quality. DNA obtained from SPE showed uniform, bright, single bands for all samples (Fig. 2). DNA obtained via SDS showed uniform, very bright bands for all samples except O. armatus Jim Porter. CHL showed faint to very faint bands for some samples, and PHE showed missing, faint, and smeared bands. By total points (Table 2), methods were ranked by the appearance of genomic DNA in the following order: SPE > SDS > CHL > PHE.
DNA was evaluated by amplifying the 1600 bp chloroplast trnS-psbC gene region for each cultivar. SPE and CHL resulted in a clear, bright band of the expected size for all samples (Fig. 3). DNA isolated from the SDS method showed minor variability in brightness, and PHE resulted in a missing band for 1 sample. By total points (Table 2), methods were ranked by the appearance of genomic DNA in the following order: SPE > CHL > SDS > PHE.
A separate experiment was carried out to examine the impact of additives to the alkaline lysis method by use of replicate samples of O. fragrans Fudingzhu and O. armatus Jim Porter. The discarding of the pellet after elution (i.e., retaining only the DNA-saturated buffer after contact with the DNA pellet and quantifying the DNA therein) decreased slightly 230 nm absorbance, and washing the pellet with ethanol significantly decreased 280 nm absorbance (Fig. 4). Treatments that included an ethanol wash increased the A260:A230 ratios by 43–59% over the control. Other additives provided no significant increase in sample purity. Addition of 2-ME, a reducing agent often added to remove tannins and other polyphenols from crude plant extracts, substantially reduced the concentration of DNA. Addition of Proteinase K, an enzyme added to the lysis buffer to digest DNAse and other contaminating proteins, led to the highest mean A260:A230 ratio (1.48 ± 0.18), although not significantly higher than washing and discarding the pellet alone (1.38 ± 0.19).
Five additional samples underwent a Wash treatment but were incubated at 4°C overnight before the DNA pellets were discarded. On average, overnight incubation increased mean DNA recovered by 100% (228 ± 146 μg vs. 485 ± 355 μg, P = 0.027; Table 5). A260:A280 was not affected by overnight incubation; A260:A230 was increased by an average of 51% (1.15 ± 0.45 vs. 1.64 ± 0.35, P = 0.004). The exclusion of the single sample with an A260:A230 “before” value of >1.5, the percent change in A260:A230 before and after overnight incubation ranged between 52 and 79%, with a mean of 63%.
The SDS method yielded DNA concentrations orders of magnitude higher than any other method and was the fastest and least expensive method. Alkaline lysis was first described for plasmid DNA extraction from bacteria and is often used for that purpose.16 In the traditional protocol, sodium hydroxide is used in the lysis buffer to denature the small, circular, double-stranded plasmid DNA into single-stranded DNA. The small, single strands of plasmid DNA then selectively reanneal whereas larger segments of bacterial genomic DNA are precipitated and removed. Dellaporta et al.18 described an alkaline lysis method for plant DNA extraction with sodium chloride replacing sodium hydroxide in the lysis buffer and additional precipitation and centrifugation steps. This method was condensed by Edwards et al.17 into single steps for precipitation and pelleting.
The Edwards method of alkaline lysis led to the recovery of large quantities of DNA in this study; however, the absorbance spectra consistently showed contamination at 230 nm, indicating the presence of carbohydrates and proteins from crude leaf extract. In the Edwards method of alkaline lysis, the DNA is pelleted in the lysis buffer without an intervening organic extraction to push cell debris and other contaminants into a separate phase from the DNA-saturated buffer. The result is a much faster extraction, but a dark-colored DNA pellet that when dissolved during elution, leads to a dark-colored solution with high 230 nm absorbance.
Additions to the Edwards method greatly improved DNA purity without sacrificing quantity, speed, or cost. Simply washing the DNA pellet in ethanol and discarding the pellet after elution led to significant improvements in DNA purity. Samples that were washed and incubated overnight before discarding the pellet showed a similar increase in purity compared with samples that were incubated 15 min before discarding the pellet, but DNA concentrations increased by an average of 100%, up to 4000 ng/μl in a 200 μl elution. One of the most attractive features of the alkaline lysis method is its speed—~1.5 h to isolate DNA from 18 samples (1.65 h with the addition of Wash). An overnight incubation greatly increases the total procedure time but may be worthwhile if the following are true: 1) the plant material is known to have very low concentrations by use of this or other methods, or 2) there is limited opportunity to perform isolation procedures (e.g., a 1-time laboratory visit or very limited supplies).
In the case of Osmanthus species, the decision regarding the DNA isolation method must be a balance of quantity, purity, and cost. SPE clearly outperformed the other methods in DNA purity by consistently maintaining the target A260:A280 and A260:A230 ratios. However, yield was low, and cost of a commercial kit was substantially higher than other methods. Methods involving CTAB lysis and organic extraction (the most traditional and widely used methods) proved highly variable in purity and concentration. One cultivar, O. armatus Jim Porter, was recalcitrant enough that isolation was unsuccessful with the use of either CTAB method.
Methods that use alkaline lysis have been used to isolate genomic DNA from a wide variety of bacterial species and plant genera, such as Nicotiana, Solanum, Glycine, Petunia, Acer, Quercus, Populus, Zea, Pinus, and Vigna (cowpea).18–22 The Edwards method of alkaline lysis led to more consistent PCR bands and larger quantities of DNA extracted from Petunia hybrida buds, leaves, and sepals as compared with CHL.21 SDS-based methods led to significantly higher yields of cowpea DNA compared with CTAB methods. An SDS-based method was recommended for cowpea DNA isolation, based on quantity and quality of isolated DNA and low cost of the method. 22
Despite the superiority of the SDS-based method versus CTAB-based methods in the current study, CTAB-based methods remain widely popular. This may be because early published approaches showed the broad usefulness of CTAB for DNA extractions, and it became ubiquitous in the literature.23 Additionally, the chemical structure of CTAB and SDS may render them more or less effective based on compounds present in plant tissue. Both CTAB and SDS are surfactants that solubilize membranes and denature proteins. SDS is a strong anionic surfactant with a hydrocarbon tail that dissolves the hydrophobic regions of proteins and a sulfate group that breaks noncovalent ionic bonds.24 CTAB is a cationic surfactant with a hydrocarbon tail and an ammonium group that are weaker denaturants compared with SDS. However, polysaccharides associated with plants are insoluble in CTAB and high concentrations of salt—a property that can be used to separate DNA effectively from plant carbohydrates.10, 11 It may be that the effectiveness of the 2 methods depends on the nature of contaminants and whether effective cell membrane solubilization and protein denaturing are more important to DNA isolation than the separation of those molecules from the DNA-containing phase. More information is needed on the secondary metabolites of species with variable performance in CTAB- versus SDS-based DNA isolation methods.
In summary, SDS is recommended for fast and cost-effective isolation of DNA from diverse Osmanthus species and cultivars. Additions to the Edwards method of Wash and the discarding of the DNA pellet after elution are recommended to increase DNA purity. With the use of this improved method, an average of 1000 μg/g DNA was isolated from fresh leaf tissue of 18 samples in ~1.5 h at a cost of 0.74 USD/sample. Rapid and cost-effective DNA isolation will serve as a foundation for molecular tool development in this genus.
The author acknowledges Carrie Witcher and Christina Jennings for invaluable technical support. Hannah Wright assisted with DNA isolation; James Bryan maintained plants. Nurseries Caroliniana was the source of all Osmanthus accessions. Two anonymous reviewers provided contributions that greatly improved the manuscript. The author has no financial support to report. Mention of a trademark, proprietary product, or vendor does not constitute a guarantee or warranty of the product by the U.S. Dept. of Agriculture and does not imply its approval to the exclusion of other products or vendors that also may be suitable. The author has no associations that may pose a conflict of interest.