Development of a method for DNA extraction from oil palm leaves and the effects of pH and ionic strength on nucleic acid quantification

The oil palm (Elaeis guineensis Jacq.) belonging to the family Arecaceae [1], a diploid oil-producing crop with a genomic size of 1.8 Gb [2], is one of the most important oil-bearing crops in the world. It is a large feather palm having a solitary columnar stem, short internodes, and short spines on both the leaf bases and within the fruit bunches [3]. It has irregular sets of leaflets on the leaf, which gives the palm its characteristic appearance. The palm is monoecious with male or female inflorescence, but hermaphroditic inflorescences sometimes develop in the axils of the leaves [3]. The fruit, which is borne on the large compact bunch, is called a drupe [1]. Distinguishing the different types of oil palms has been controversial. These attempts have been unsatisfactory since in the wild state, each palm represents a hybrid with respect to some of its traits [1]. Oil palm is classified based on the fruit type and fruit form. It has three fruit forms: Dura, Pisifera and Tenera (hybrid fruit form) and different fruit types namely, virescens, albescens, nigrescens and poissoni [1]. Oil palm has benefited immensely from conventional breeding program in Nigeria. This has solely been made possible through the dedicated breeding program put in place by the Nigerian Institute for Oil Palm Research (NIFOR). The progress made so far has been very limited for two reasons: (1) the long generation time and (2) the outcrossing nature of the crop. With the emergence of deoxyribonucleic acid (DNA) marker technologies, scientists see the possibility that significant success can be achieved if markers are extensively applied by oil palm breeders as exemplified by the cloning of the shell thickness gene [4]. Exploration of DNA marker technologies, combining the knowledge from research in molecular genetics and genomics, offers great possibilities to oil palm breeding [5]. With DNA marker technologies, the underlying genetic basis of phenotypic traits can be studied independent of environmental influences. However, critical to the adoption, application and the domestication of these technologies is an effective DNA extraction method that is less complex than the methods that have been previously applied to extract DNA from palms [6-10]. The basic principles underlying DNA extraction procedures are not very complicated, but the growing numbers of DNA extraction procedures indicate that it is not always simple and the published protocols are not necessarily reproducible for all species [11,12]. The objective of this study was to develop a method for DNA extraction from oil palm leaves that is cost effective and adaptable to low budget laboratories.

A protocol for the extraction of total cellular nucleic acids from leaf tissue of oil palm was embarked upon without the use of liquid nitrogen.

To determine the purity and quantity of extracted nucleic acids, four different solvents (sterile distilled water, deionized water, diluted TE buffer (1:2 with distilled water) and TE buffer) were used. The quality of the extracted nucleic acids was also verified by separation on 0.8% to 1% agarose gel stained with ethidium bromide. The nucleic acid concentrations (ng/µl) and the optical density (OD) ratio A_260/280 were calculated by measuring the OD at wavelength of 260 nm and 280 nm using a spectrophotometer (Genesys 10uv Scanning, Thermo Electron Corporation, USA: model Genesys 10-S).

To determine the usability of the extracted DNA for PCR, primer combinations were designed from three simple sequence repeats (SSR) sequences mEgCIR3275, mEgCIR3533 and mEgCIR3557 [10] using the online program Primer3 (http://frodo.wi.mit.edu/):

mEgCIR3275M13for 5’-TTTCCCAGTCACGACGTTGGTGGAAGCTTTTTGTCTGC-3’and mEgCIR3275rev 5’-ATTGAAGAGGGCAGGGTTTT-3

mEgCIR3533M13for 5’-TTTCCCAGTCACGACGTTACGGTCTATGGCTCTGTCGT-3’and mEgCIR3533rev 5’-ACATGAGGAAAGCGCTAGGA-3’

mEgCIR3557M13for 5’-TTTCCCAGTCACGACGTTCATTGCCATTCCCTTCAAGT-3’and mEgCIR3557rev 5’-TCCCCTCTGTTCACTCAAGC-3’.

Primers were labeled using M13 tailing procedure (shown in bold) [13].To visualize the PCR products, anIRD-800(=DY-781) labeled M13 primer was added to each SSR primer combination. All unlabeled primers were ordered from Life Technologies GmbH, Darmstadt, Germany, the M13-IRD800 (=DY-781) labeled primers were from Biomers, Ulm, Germany.

The PCR amplification was performed in a GeneAmp^® PCR System 2700 (Applied Biosystems) thermocycler. For the PCR, 4 µl of DNA (50 ng/µl) was mixed with 11 µl master mix. The master mix contained 0.3 µl dNTPs (10 mM), 8.7 µl H₂O, 1.5 µl 10 × PCR buffer, 0.15 µl M13-IRD800 primer (5 pmol/µl) for labeling, 0.05 µl FIREPol DNA polymerase (5 U/µl, Solis Biodyne), 0.15 µl forward primer (5 pmol/µl) and 0.15 µl reverse primer (5 pmol/µl).

Statistical analysis was carried out using GenStat version 8.1. Duncan Multiple Range Test (DMRT) was used for all pairwise comparison of means at 95% probability.

In the present study, a new protocol for nucleic acid extraction from oil palm leaves was developed without the use of liquid nitrogen (Fig. 1). This method isolates DNA, which can be well amplified by PCR, even though the nucleic acid preparations will contain traces of RNA. No differences were observed in the total DNA extracted from the three fruit forms (Pisifera, Dura and Tenera) of oil palm E. guineensis in terms of size. The method reported here was also used to extract DNA from a large population of oil palm (93 trees) (Fig. 2). The method is considered fast as it takes only about 25 min to complete one extraction and it is considered easy as it involves only seven steps. The DNA separated on 1% agarose gel was of high quality with a major band larger than 23 kb with no visible sign of degradation by DNase. The extracted DNA samples dissolved in TE buffer were stable for more than a year, when stored at −20ºC. The minimum storage period was 90 days before PCR amplification. In some instances, extracted DNA was stored for 1 year prior to PCR amplification. The key steps in the developed DNA extraction protocol are the fast homogenization of the leaf tissue in the presence of DNA extraction buffer using mortar and pestle, and the introduction of 5 M potassium acetate before the phenol:chloroform:isoamylalcohol (25:24:1) step. Potassium acetate is known to form complexes with proteins and polysaccharides [12,14]. Proteins and polysaccharides are often the major contaminants during DNA extraction because they form complexes with DNA and they must be eliminated to enhance the purity of the DNA [14]. Proteins are known to interfere with enzymes activities in both restriction digestions and polymerase reactions [15]. The ability of the DNA polymerase to selectively amplify the DNA extracted with the developed protocol means that the DNA is good enough for PCR reaction and also inferred that it can be used for restriction digestions. Also, the presence of residual RNA in the sample (as the method presented here extracted total cellular nucleic acids) does not present any problem to the use of the extracted DNA for molecular analyses as shown in Figure 3. This agrees with the findings of Murray and Thompson [15], thus eliminating the need for RNase treatment of the DNA, which would be an extra cost for a low budget laboratory. Diluted DNA solution (5 ng/µl, using double distilled sterile water) for PCR reactions, also stored at -20ºC, gave reproducible products over at least a three-month period (Fig. 3).

The comparison made between the different DNA extraction methods previously applied in palms and the new method presented in this study is shown in Table 1. The components of the DNA extraction buffer used in the new protocol are entirely new and in several ways different from a previous protocol used for mature leaves of oil palm [6]. Mannitol, PEG 6000, BSA and β-mercaptoethanol (to mention a few), which formed part of the extraction buffer of Ying and Zaman [6], are not part of the new extraction buffer reported here and the concentrations of Tris-HCl, EDTA and SDS are also different. The method reported here is also very different from the methods described for date palm [7,8]. Liquid nitrogen, which is not always easy to obtain in most third world laboratories, was used by the methods of Risterucci et al. [9] and Billotte et al. [10] (Table 1). The cost of DNA extraction per sample in Nigeria using the new method is only a third of the most expensive protocol [9,10] and considerably less expensive than any of the published methods[6,7,8] as shown in Table 1. Regarding the leaf sampling for DNA extractions, three statements can be made: 1) DNA extraction worked best with fresh leaf samples prior to any drying procedure, 2) the DNA extraction method is non-destructive to either mature or nursery palms, and 3) oven-dried leaf sample (at 50ºC) and very young (spear) leaves can also be used for DNA extraction.

Following the successful extraction of DNA from any sample, the next most important steps are the determination of the purity and concentration of the extracted DNA. Here, we determined the effect of different solvents (sterile distilled water, deionized water, diluted TE buffer [1:2 with distilled water] and TE buffer) on the spectrophotometric absorbance of the extracted nucleic acids. The method reported here extracted total cellular nucleic acids comprising of DNA and RNA as visible in Figure 1. The treatment of the nucleic acids with RNase was omitted for two reasons. First, the cost of RNase in Sub-Saharan Africa is very high and cost is one of the major factors considered in this work. Secondly, the presence of RNA did not interfere with the PCR amplification of the extracted DNA. Thus the values obtained at A₂₆₀ actually represent DNA and RNA. From our result, the nucleic acids diluted with sterile distilled water had the highest absorbance value of 0.143 at A₂₆₀ followed by that of deionized water (0.137), diluted TE buffer (0.129) and TE buffer (0.124), respectively (Table 2). These values were significantly different (P ˂ 0.05). These differences in the mean value of A₂₆₀ using the same nucleic acid extract but different solutions for dilution may be due to the ionic strength and the pH of the solutions. This agrees with the report of Wilfinger et al. [16], who reported that RNA absorbance at A₂₆₀ was dependent on the pH and the ionic strength of the solution used for spectrophotometric analysis. They further stated that the pH and ionic strength of the solution can substantially influence the qualitative and quantitative determination of nucleic acids as observed in our study.

It is established that resonance structure of the pyrimidine and purine bases are responsible for the maximal absorption of nucleic acids and proteins in the 260–280 nm regions of the UV spectra [16]. Molecular biology manuals [14,18,19] report that an A_260/280 ratio of 1.8 is an indication for pure DNA preparations and an A_260/280 ratio of 2.0 for pure RNA preparations free of proteins. The absorption spectra for nucleic acid bases, nucleosides and nucleotides are reported to be strongly pH-dependent because of the degree of ionization of the bases at different pH values [17]. This information has not been adequately emphasized in molecular biology reports [16]. In our findings, A_260/280 ratio indicated the presence of DNA in the sample diluted with sterile distilled water (1.987 ± 0.029) against the other solutions, which indicated more of RNA than DNA at a value of 2.005 ± 0.010 (for deionized water), 2.087 ± 0.023 (for 1:2 diluted TE buffer) and 2.078 ± 0.006 (for TE buffer). The values obtained for the different water solutions were significantly different from the values obtained for the TE buffers (P ˂ 0.05), but were not significantly different when compared within, that is, sterile distilled water versus deionized water, and 1:2 diluted TE buffer versus TE buffer. These values increased from the different water samples to the different TE buffers and they were directly related to the increase in the pH and conductivity of the dilution solutions (Table 2). This finding agrees with the findings of Wilfinger et al. [16].

In our experiments, we observed that changes in the pH and ionic strength of the different solutions used for the determination of A_260/280 ratio of the same extracted nucleic acid sample altered the A_260/280 ratios (Table 2) and this agrees with the findings of Wilfinger et al. [16] reported for RNA preparation. Ratio greater than 2.0 is often recorded for A_260/280. This may be related to DNA degradation and measurement of free nucleotides, but this assumption is not applicable in our case because the same DNA sample that gave an A_260/280 value less than 2.0 for sterile distilled water also gave a value greater than 2.0 value for 1:2 diluted TE and TE buffers by simply altering the pH and ionic strength of the solutions. It can be stated from our results that the apparent A_260/280 ratio varies from one solvent to the other due to the pH and ionic strength of the different solutions. Proteins absorb light at 280 nm and 230 nm. Thus, A_260/230 ratio could also be used for the detection of protein contaminations, but this is rarely done because common buffers such as Tris are known to also absorb light at 230 nm as well as phenol and other organic contaminants. The widely used ratio for determination or assessing DNA purity is A_260/280, which is easy to determine. DNA absorbs light so strongly at 260 nm and it takes significant protein contamination to have a noticeable effect on the A_260/280 ratio.

We also compared the effects of pH and ionic strength on nucleic acid concentration and yield using different solvents. The highest nucleic acid yield of 460.8 µg was obtained for a sample diluted in sterile distilled water. 438.4 µg was obtained in deionized water and 413.9 µg and 396.8 µg was obtained in 1:2 diluted TE buffer and undiluted TE buffer, respectively. The mean yields were significantly different (P ˂ 0.05) from each other except for 1:2 diluted TE and undiluted TE buffer (Table 2; column 7). Similarly, the mean nucleic acid concentration was inversely correlated with the pH and conductivity of the solvent and the values were statistically significant for all four solvent conditions (P < 0.05) (Table 2; column 6). Our results suggest the nucleic acid yield and concentration were inversely related to the pH and conductivity of the solution used. These findings are consistent with the findings from Wilfinger et al. [16], who reported the same observations for RNA by comparing different solvents.

In conclusion, a high throughput nucleic acid extraction method for oil palm was developed that results in amplifiable DNA and allows fast handling of large populations and is cost-effective. Separation of the nucleic acids on a 1% agarose gel showed the presence of high quality DNA with one major band larger than 23 kb with no visible signs of degradation by DNase. RNA present in the lower part of the gel that was extracted together with the DNA does not present a problem for its use in PCR reactions due to the substrate specificity of the DNA polymerase and therefore eliminates the necessity for RNase treatment. The absorbance values obtained from the spectrophotometric evaluation of the nucleic acid extractions represent total cellular nucleic acids because both DNA and RNA are known to absorb at 260 nm. The omission of an RNase treatment in the extraction process will lead to an over-estimation of these values for the DNA. For laboratories that can afford RNase and desire accurate evaluation of DNA concentration using this method, the extracted nucleic acids can further be treated with RNase following standard procedure [14]. The yields and concentrations of the extracted nucleic acids were subject to the variation in pH and ionic strength of the solvents used for the measurement of photometric absorbance. However, the accuracy of the nucleic acid measurement, even though it is only a rough estimate, is sufficient in its precision to perform PCR for qualitative evaluation of bands as required in marker-assisted selection for breeding programs.

We are grateful to the Executive Director of the Nigerian Institute for Oil Palm Research (NIFOR) for his permission to publish this work and also to the Agricultural Research Council of Nigeria (ARCN) for support. We are also grateful to the Institut für Biowissenschaften, Pflanzengenetik, Universität Rostock for hosting part of this research work. We like to thank Dr. Gupta Jagadis Kapuganti for correcting the manuscript. The authors declare that they have no competing interest.